RSeC
'Theopen University
U
ROYAL SOCIETY OF CHEMISTRY
The
Molecular World
Metals and Chemical Change
This publication forms part of an Open University course, S205 The Molecular World. Most of the texts which make up this course are shown opposite. Details of this and other Open University courses can be obtained from the Call Centre, PO Box 724, The Open University, Milton Keynes MK7 6 Z S , United Kingdom: tel. +44 (0)1908 653231, e-mail
[email protected] Alternatively, you may visit the Open University website at http://www.open.ac.uk where you can learn more about the wide range of courses and packs offered at all levels by The Open University. The Open University, Walton Hall, Milton Keynes, MK7 6AA First published 2002 Copyright 0 2002 The Open University All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, transmitted or utilized in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without written permission from the publisher or a licence from the Copyright Licensing Agency Ltd. Details of such licences (for reprographic reproduction) may be obtained from the Copyright Licensing Agency Ltd of 90 Tottenham Court Road, London W l P OLP. Edited, designed and typeset by The Open University. Published by the Royal Society of Chemistry, Thomas Graham House, Science Park, Milton Road, Cambridge CB4 OWF, UK. Printed in the United Kingdom by Bath Press Colourbooks, Glasgow. ISBN 0 85404 665 8 A catalogue record for this book is available from the British Library. 1.1 s205book 4 i 1 . 1
This series provides a broad foundation in chemistry, introducing its fundamental ideas, principles and techniques, and also demonstrating the central role of chemistry in science and the importance of a molecular approach in biology and the Earth sciences. Each title is attractively presented and illustrated in full colour.
The Molecular World aims to develop an integrated approach, with major themes and concepts in organic, inorganic and physical chemistry, set in the context of chemistry as a whole. The examples given illustrate both the application of chemistry in the natural world and its importance in industry. Case studies, written by acknowledged experts in the field, are used to show how chemistry impinges on topics of social and scientific interest, such as polymers, batteries, catalysis, liquid crystals and forensic science. Interactive multimedia CD-ROMs are included throughout, covering a range of topics such as molecular structures, reaction sequences, spectra and molecular modelling. Electronic questions facilitating revisiodconsolidation are also used.
The series has been devised as the course material for the Open University Course S205 The Molecular World. Details of this and other Open University courses can be obtained from the Course Information and Advice Centre, PO Box 724, The Open University, Milton Keynes MK7 6ZS, UK; Tel+44 (0)1908 653231; e-mail:
[email protected]. Alternatively, the website at www.open.ac.uk gives more information about the wide range of courses and packs offered at all levels by The Open University.
Further information about this series is available at www.rsc.org/rnoEecuEarworld. Orders and enquiries should be sent to: Sales and Customer Care Department, Royal Society of Chemistry, Thomas Graham House, Science Park, Milton Road, Cambridge, CB4 OWF, UK Tel: +44 (0)1223 432360; Fax: +44 (0)1223 426017; e-mail: sales @rsc.org
The titles in The Molecular World series are: edited by Lesley Smart and Michael Gagan edited by David Johnson edited by Michael Mortimer and Peter Taylor edited by Elaine Moore edited by Peter Taylor and Michael Gagan edited by Lesley Smart edited by Charles Harding, David Johnson and Rob Janes edited by Peter Taylor
Course Team Chair Lesley Smart
Tim Martin Jessica Barrington
Open University Authors Eleanor Crabb (Book 8) Michael Gagan (Book 3 and Book 7) Charles Harding (Book 9) Rob Janes (Book 9) David Johnson (Book 2, Book 4 and Book 9) Elaine Moore (Book 6) Michael Mortimer (Book 5 ) Lesley Smart (Book 1, Book 3 and Book 8) Peter Taylor (Book 5, Book 7 and Book 10) Judy Thomas (Study File) Ruth Williams (skills, assessment questions) Other authors whose previous contributions to the earlier courses S246 and S247 have been invaluable in the preparation of this course: Tim Allott, Alan Bassindale, Stuart Bennett, Keith Bolton, John Coyle, John Emsley, Jim Iley, Ray Jones, Joan Mason, Peter Morrod, Jane Nelson, Malcolm Rose, Richard Taylor, Kiki Warr.
Course Reader Cliff Ludman
Course Manager Mike Bullivant Course Team Assistant Debbie Gingell Course Editors Ian Nuttall Bina Sharma Dick Sharp Peter Twomey CD-ROM Production Andrew Bertie Greg Black Matthew Brown Philip Butcher Chris Denham Spencer Harben Peter Mitton David Palmer BBC Rosalind Bain Stephen Haggard Melanie Heath Darren Wycherley
Course Assessor Professor Eddie Abel, University of Exeter Audio and Audiovisual recording Kirsten Hintner Andrew Rix Design Steve Best Carl Gibbard Sarah Hack Mike Levers Sian Lewis John Taylor Howie Twiner Library Judy Thomas Picture Researchers Lydia Eaton Deana Plummer Technical Assistance Brandon Cook Pravin Pate1 Consultant Authors Ronald Dell (Case Study: Batteries and Fuel Cells) Adrian Dobbs (Book 8 and Book 10) Chris Falshaw (Book 10) Andrew Galwey (Case Study: Acid Rain) Guy Grant (Case Study: Molecular Modelling) Alan Heaton (Case Study: Industrial Organic Chemistry, Case Study: Industrial Inorganic Chemistry) Bob Hill (Case Study: Polymers and Gels) Roger Hill (Book 10) Anya Hunt (Case Study: Forensic Science) Corrie Irnrie (Case Study: Liquid Crystals) Clive McKee (Book 5 ) Bob Murray (Study File, Book 11) Andrew Platt (Case Study: Forensic Science) Ray Wallace (Study File, Book 11) Craig Williams (Case Study: Zeolites)
METALS AND CHEMICAL CHANGE David Johnson and Kiki Warr
1.1 Metals and their physical properties
13
1.2 Summary of Section 1
16
2.1 Oxidation and reduction
18
2.2 Oxidation of metals by aqueous hydrogen ions
20
2.3 Reactions of metals with aqueous metal ions
23
2.4
Reactions of metals with oxygen and the halogens
25
2.5
Summary of Section 2
26
3.1 Mercury
28
3.2 Tin
31
3.3 Aluminium
32
3.4 Summary of Section 3
36
4.1
A critical look at the hypothesis
38
4.2
Summary of Section 4
40
6.1 Summary of Sections 5 and 6
44
7.1 Entropy
46
7.2 The direction of heat flow
47
7.3 The direction of chemical change
48
7.4 Summary of Section 7
51
8.1 The enthalpy change for a pure substance
52
8.2 Measuring enthalpy changes: calorimetry
56
8.3 Molar enthalpy changes
60
8.4 Summary of Section 8
62
9.1 Hess’s law
63
9.2 Standard enthalpy changes
64
9.3 Standard enthalpies of formation 9.3.1 Enthalpies of formation of pure substances 9.3.2 Calculating standard enthalpies of reaction 9.3.3 Standard enthalpies of formation of aqueous ions
65 65 66 68
9.4 Summary of Section 9
70
10.1 Determining entropy changes
71
10.2 Entropy changes for phase transitions
71
10.3 Entropy changes of substances with temperature
73
10.4 Absolute entropies: the third law of thermodynamics
79
10.5 Absolute entropy of chlorine gas at 298.15 K
80
10.6 Absolute entropies in the Data Book
82
10.7 The entropy change for a reaction
82
10.8 Summary of Section 10
83
11.1 The Gibbs function and the equilibrium constant
85
11.2 Final survey of the thermodynamic database
87
11.3 Summary of Section 11
90
12.1 Metals and their aqueous ions
92
12.2 Summary of Section 12
96
13.1 Summary of Section 13
101
15.1 Some consequences of Table 15.1 15.1.1 Ball lightning
106 106
15.2 Summary of Sections 14 and 15
108
16.1 Reactions of solids and gases: the sign of AS:
112
16.2 Summary of Section 16
113
17.1 The variation of AG; with temperature
115
17.2 Carbon as a reducing agent
118
17.3 A survey of metal extraction methods
119
17.4 Summary of Section 17
121
18.1 Comparing sodium and silver halides
125
18.2 Summary of Section 18
126
20.1 The energy of interaction between two ions
130
20.2 The lattice energy of an ionic crystal
132
20.3 Applying the Born-Land6 equation
134
20.4 The Kapustinskii equation
135
20.5 A more general Born-Haber cycle
136
20.6 Summary of Section 20
138
22.1 Changes in ionization energy down a Group
146
22.2 Summary of Section 22
147
23.1 The alkali metals 23.1.1 The alkali metals in liquid ammonia
149 151
23.2 Summary of Section 23
153
24.1 Manufacture of sodium hydroxide and chlorine 24.1.1 The membrane cell
155 156
24.2 Uses of sodium hydroxide and sodium carbonate
158
24.3 Summary of Section 24
159
25.1 Alkali metal halides
160
25.2 Oxygen compounds of the alkali metals 25.2.1 The relative stability of peroxides and oxides
161 163
25.3 Hydrides and nitrides of the alkali metals
167
25.4 Summary of Section 25
168
26.1 Metal complexes
170
26.2 The anatomy of a complex
173
26.3 Summary of Section 26
175
27.1 Alkali metal anions
180
27.2 Summary of Section 27
183
28.1 The alkaline earth metals
185
28.2 Why are there no Group I1 monohalides?
188
28.3 Group I1 in industry: lime and its applications
192
28.4 Oxides, hydroxides and carbonates of the Group I1 elements
197
28.5 Are the Group I1 dihalides ionic?
198
28.6 Complexes of Group I1 elements
200
28.7 Summary of Section 28
201
CASE STUDY: BATTERIES AND FUEL CELLS Ronald Dell and David Johnson
5.1 Primary batteries
247
5.2 Secondary batteries 5.2.1 Lead-acid batteries 5.2.2 Alkaline batteries 5.2.3 Lithium batteries
250
250 25 1 252
M e t a l s and Chemical C h a n g e
In the earlier books in this series, there has been an emphasis on molecular and electronic structure -that is, on the spatial arrangement of atoms within chemical substances, and on the arrangement of electrons within atoms. Very little has been said about chemical change. But here this emphasis shifts, and we ask why chemical reactions happen. There are two conditions that must be fulfilled before a chemical reaction can occur: the equilibrium constant must be sufficiently favourable, and the rate must be sufficiently fast. This Book will be concerned with both conditions, but mainly with the first. You will meet new ‘labour-saving’ properties of chemical substances; these will allow us to predict whether a chemical reaction has a favourable equilibrium constant or not; the reaction does not even have to be tried out. The units of the properties in question are mainly those of energy, and come from a branch of science called thermodynamics. To show the relevance of this subject, we shall use it to explore an important problem about the chemical behaviour of metals. Finally, when our study of this problem is complete, thermodynamics is used again, towards the end of the Book, in a systematic study of the chemistry of the alkali and alkaline earth elements -that is, of Groups I and I1 of the Periodic Table. Along with thermodynamics, metals are therefore a major theme in this Book, so we begin by reminding you about them, and about the way that their properties are explained by the simplest theory of metallic bonding.
Figure 1.1 shows a full Periodic Table, colour-coded to reveal the periodic distribution of metals, semi-metals and non-metals. Of the 114 known elements, 90 are, or are likely to be, metallic. This, and the other books in the series, concentrate on the 46 typical elements. Here, metals are not so predominant, but, even so, they still outnumber each of the other two categories. Some metallic elements, such as bismuth and manganese, are brittle, but most, when pure, are malleable and ductile. Malleable materials are those that can be reshaped by hammering; ductile materials can be drawn out under tension into wires. Figure 1.2 shows a piece of early British gold jewellery dated 1600 BC. Such things were made by hammering out gold into sheets. This is possible because the metal is malleable, and at the same time strong. Malleability and ductility are especially associated with those metals that possess one of the two close-packed structures discussed in The Third Dimension: Crystals1*. What are the names of these two types? Hexagonal close-packed and cubic close-packed.
* See the references in Further Reading (p. 234) for details of other titles in The Molecular World series that are relevant. 13
Figure 1.1 Long form of the Periodic Table, showing metals, semi-metals and non-metals. Isotopes of elements 110-112,114and 116havebeen made, but these elements have so far [2002] not been given names.
Figure 1.2 Gold lanula from Gwynedd, ca. 1600 BC.
14
Both close-packed structures consist of layers of atoms of the metallic element. It follows that if we can explain why such layers can slip over one another fairly easily, we can account for both malleability and ductility. Now a simple model of a metal consists of a regular array of positively charged ions in a ‘pool’ of freely moving electrons. The interaction between the positive ions and the negatively charged electrons, which surround the ions, holds the metal together. Let us contrast the situations in a metal and in an ionic solid, such as NaC1, when the layers are displaced relative to one another. Look first at Figure 1.3b. Why should such a displacement be unfavourable in an ionic solid? The displacement brings like charges in adjacent layers into close proximity. Repulsion between the charges will then push the layers apart. This explains why fracture and not deformation is usually the result of beating an ionic solid. The contrast with the situation in a metal (Figure 1.3a) is obvious: all the ions are of like charge, the situation after displacement is similar to what it was before, and the freely moving electrons can adjust to the change without further disruption. Consequently, in a pure metallic crystal, layers can usually slip easily over one another, thus accounting for the properties of ductility and malleability.
Figure 1.3 Two different types of solid undergoing a shearing movement along a plane (red broken line) through the crystal: (a) the crystal of a metal; (b>ithe crystal of an ionic solid contaming positive and negative ions.
You will be familiar with other properties of metals from your everyday contact with iron, aluminium, copper, silver and tin, for example. They often have a lustrous appearance, and are good conductors of heat. However, the most characteristic property of metals is their high electrical conductivity. This is explained by the free electrons that roam at random through the metallic structure. When a voltage difference is applied across two points on a piece of metal, the motion of the electrons becomes less random, there is an overall movement of electrons between the two points, and an electric current flows. The unit of electrical conductivity is siemens per metre (S m-’) *. Those elements classified as metals in Figure 1.1 have an electrical conductivity at or below room along any direction in a single crystal of any temperature of at least 3 X lo5 s known form of the element. Although it is a good electrical conductor, carbon in the
* The electrical conductivity in siemens ~
‘
p
~
kin ~
~
e
volt is placed two opposite faces of a cube of the material with sides of one metre.
15
~
form of graphite does not meet this criterion. The structure of graphite is shown in Figure 1.4. It consists of sheets of carbon atoms, and each atom is bonded to three others in the same sheet. Carbon has four outer electrons in the shell structure (2,4); in graphite, each carbon atom shares three of these with three other carbon atoms in the same sheet by forming three covalent C-C bonds. The fourth electron is mobile, just as the bonding electrons in a metal are mobile; it binds carbon atoms within its sheet more strongly together by contributing to a pool of electrons concentrated around the plane of the sheet. Consequently, there is high conductivity parallel to the sheets, but very low conductivity at right-angles to them.
Figure 1.4 The structure of graphite; there is metallic conductivity in directions parallel to the sheets of carbon atoms, but not ai. right-angles to them.
Because of this property, graphite is sometimes called a two-dimensional metal. However, our decision to treat as metals only those elements with high electrical conductivity in all three dimensions rules graphite out, and carbon is not classified as a metal in Figure 1.1.
1 2
3
16
Most of the chemical elements are metals, and even among the typical elements, metals predominate. Most, but not all, metallic elements are malleable and ductile. Elements classified as metals have high electrical conductivities in all directions in a single crystal of the element. Using this criterion, carbon is a non-metal. These properties can be explained by a model in which metals are regarded as positive ions immersed in a pool of free electrons.
Between now and the end of this Book, we shall investigate an important problem about the chemical behaviour of metals. Certain features of the problem are familiar to everybody. Gold jewellery (Figure 2.1) can survive essentially unchanged for thousands of years. Many bronze busts and statues (Figure 2.2) are much more recent, but the green stains on their stone bases show that significant corrosion of their copper content has already taken place. Again, the uranium metal intended for Nazi Germany’s first nuclear reactor went up in flames when a physicist took a shovel to it; rubidium reacts violently with water, and inflames in air, without the assistance of a shovel (Figure 2.3). Few chemists would quarrel with you if you said that copper was more reactive than gold, and that rubidium was more reactive than uranium, but what exactly do we mean by the word ‘reactive’? Again, is our statement about reactivities true only in moist air, or does it remain correct in the presence of other chemicals? It is these and other questions that we shall examine for the remainder of this Book. To begin with, you must learn a little more about the reactions of metals. The next three Sections give you the opportunity to do this, partly by reading, and partly by watching experiments on the CD-ROM application, Reactions of metals.
Figure 2.1 Gold torque from Stanton, Staffordshire, ca. 1OOOBC.
Figure 2.2: Bronze is about 90% copper and 10% tin‘; the green stains on the base of this statue of Scotland’s hero, William Wallace, are an indication of copper corrosion.
Figure 2.3 Rubidium reacting with water and air. 17
This activity introduces the problem of metal reactivity, and then asks you to record what happens when the metals copper, iron, magnesium, tin and zinc are added to dilute acid. It is the first sequence of Reactions of metals on the CD-ROM associated with this Book. As noted in the latter part of Activity 2.1, in those cases where a reaction happened, the metal dissolved in the dilute acid to form a dipositive aqueous ion. For example, in the case of iron, this ion is Fe2+(aq).Moreover, when a reaction occurred, bubbles of gas were evolved. This happened for iron, magnesium, tin and zinc. You may remember that magnesium reacts with acids to yield hydrogen gas. Thus, for iron, a likely equation for the reaction is: Fe(s) + 2H+(aq) = Fe2+(aq) + H2(g) (2.1) Notice that this equation has been pared down to just the chemical species that change during the reaction; species like the chloride ion, C1-(aq), which are present but do not change, have been left out. Try to write equations for the other three reactions that you observed. Following the example of iron in Equation 2.1, you should get: Mg(s) + 2H+(aq) = Mg2+(aq) + H2(g> Sn(s) + 2H+(aq) = Sn2+(aq)+ H2(g) Zn(s)
+ 2H+(aq) = Zn2+(aq)+ H2(g)
In Reactions 2.1-2.4, you can see one of the most prominent chemical characteristics of metals. What is it? They form positive rather than negative ions in aqueous solution. Metals also tend to form positive ions in solid compounds. For example, we say that solid sodium chloride contains Na+ ions, and this contrasts with the behaviour of non-metals, which do not form positive ions, and indeed often form monatomic negative ions such as C1- or 02-.There are important terms for describing changes of this type, to which we now turn.
The words oxidation and reduction tend to take on broader meaning as one learns more chemistry. You are now ready to take a step down this road. You know that a substance is said to be oxidized when it reacts with oxygen. Thus, magnesium is oxidized when it burns in oxygen gas (Figure 2.4):
Let us consider how electrons are redistributed during this reaction by using an ionic bonding model.
18
Figure 2.4
Magnesium metal burning in the oxygen of the air.
How are magnesium and oxygen held together in MgO? Each magnesium atom loses two electrons and forms the ion Mg2+,which has the electronic structure of neon; each oxygen atom gains two electrons and forms the ion 02-,which also has the electronic structure of neon. The forces between the oppositely charged ions in Mg2+02-hold the compound together. Thus, when the magnesium atoms on the left of Equation 2.5 are oxidized, they lose electrons and become Mg2+ions in MgO. Chemists fasten on to this electronic change, and use it in a broader definition of oxidation:
Another feature of Equation 2.5 is that the oxygen atoms, bound together on the left-hand side as diatomic molecules, take on electrons and form 02-ions. Thus, whereas the magnesium atoms lose electrons, the oxygen atoms gain them. The taking-up of electrons, a process that is the reverse of oxidation, is called reduction:
A useful mnemonic that will help you to remember this is OILRIG: Oxidation Is Loss; Reduction Is Gain.
19
In Equation 2.5, notice that oxidation and reduction occur together; they are complementary processes: the magnesium is oxidized and the oxygen is reduced. Apart from certain exceptional cases, a reaction that includes oxidation, also includes reduction, and vice versa. To mark this fact, such reactions are often called redox reactions, and Reactions 2.1-2.4 are typical examples. In Equations 2.1-2.4, is the metal oxidized or reduced? The metal loses electrons and forms positive ions; it has been oxidized. In Equations 2.1-2.4, what happens to the hydrogen ions? They are reduced; they gain electrons and form neutral hydrogen atoms, which are combined in diatomic molecules.
Which of the following are redox reactions? In each redox reaction, identify the element that is oxidized, and the element that is reduced: (i) K(s) + H+(aq) = K+(aq) + iH,(g) (ii) Cu2+(aq)+ Fe(s) = Cu(s) + Fe2+(aq) (iii)Mg(s) + F2(g) = MgF2(s) (iv) Ca2+(aq)+ 2F-(aq) = CaF2(s) (v) Cl,(g) + 2Fe2+(aq)= 2Cl-(aq) + 2Fe3+(aq)
We have now established that Reactions 2.1-2.4 are all redox reactions. Furthermore, in all of them, a metal atom reacts with two hydrogen ions to give an ion with two positive charges and a molecule of hydrogen gas. However, Activity 2.1 revealed a significant difference in the vigour of the reactions. In particular, the magnesium reaction was quite violent; by contrast, the iron, tin and zinc reactions were slow. Indeed, the very slow tin reaction only became perceptible on heating. Similar reactions, which are as violent or more violent than the magnesium reaction, are observed when the alkali metals (Group I of the Periodic Table) react with acids. For example, steady evolution of hydrogen occurs from the surface of lithium metal, but with sodium the gas evolution is extremely vigorous; for potassium, rubidium and caesium, the reaction is explosive, and the hydrogen gas catches fire. The reactions with plain water are very similar.
20
Write equations for the reactions of lithium and caesium with acids, in which hydrogen gas is produced. Like magnesium, the alkali metals form ions with noble gas structures, losing one electron to form singly charged cations:
iH2(g)
Li(s)
+ H+(aq) = Li+(aq) +
Cs(s)
+ H+(aq) = Cs+(aq) + iH,(g)
(2.6)
So experiments show that magnesium and the alkali metals react much more violently with aqueous hydrogen ions than do zinc, iron and tin. However, in Activity 2.1 there was one element that did not react at all with dilute hydrochloric acid.
Which element was this? It was copper; when copper is dropped into dilute hydrochloric acid, no hydrogen gas is evolved, and the dipositive aqueous ion, Cu2+(aq),is not formed. Now hydrated forms of copper sulfate, CuS04, are used by gardeners as a fungicide. The solution of this compound in water is blue because of the blue ion Cu2+(aq) (Figure 2.5): CuS04(s) = Cu2+(aq)+ S042-(aq) (2.8) Thus, the ion Cu2+(aq)exists.
Figure 2.5 Dried copper sulfate, CuS04 (left) is colourless, but it dissolves in water to form the blue aqueous ion, Cu2+(aq)(centre). On standing in moist air, dried CuS04 absorbs water and turns blue, forming the solid copper sulphate used by gardeners (right). This has the empirical formula Cu S04.5H20.
Write an equation for a reaction of metallic copper with acid to produce hydrogen gas. Cu(s) + 2H+(aq) = Cu2+(aq)+ H2(g) As Activity 2.1 has demonstrated, this reaction does not happen. This is true of the analogous reaction for certain other metals, in particular silver and gold. Silver forms a well-defined colourless ion Ag+(aq) and, for our purposes, the most convenient description of aqueous oxidized gold is the ion Au3+(aq).Thus, we can write two further equations for reactions that do not occur.
21
Write them down now! (2.10) Au(s)
+ 3H+(aq) = Au3+(aq) +
Hz(g)
(2.1 1)
We now have three metals, copper, silver and gold, which we know can form ions under some circumstances, but cannot do so by a reaction with hydrogen ions. Hydrogen ions can oxidize the alkali metals, magnesium, zinc, iron and tin, but
not copper, silver and gold. Let us now summarize the observations made in this Section. We can do it by making what seems a somewhat arbitrary classification of the metals that we have discussed, but the classification will later turn out to be useful. In the first class are those metals that are not oxidized by aqueous hydrogen ions at all. Important examples are copper, silver and gold. In the second class we shall put metals, such as zinc, which react mildly with hydrogen ions. Iron, which forms a dipositive ion Fe2+(aq)when it reacts with dilute acid, is also a member of this class. Finally, we are left with a third class, the alkali metals and alkaline earth metals (magnesium, calcium, strontium and barium), which react violently with hydrogen ions. Examples of the behaviour of the three classes with hydrogen ions are illustrated in Figure 2.6.
Figure 2.6 Differences in the vigour of the reaction of metals with hydrogen ions exemplified by (a) copper, (b) zinc, and (c) potassium.
Now, all the metals that we have considered have been subjected to the same test: an attempt to oxidize them with aqueous hydrogen ions. Suggest a grading of the three classes in order of the tendency of the metals to be oxidized to aqueous ions by H+(aq). A reasonable hypothesis is as follows: those metals that are unaffected by dilute acids (the first class) are the least readily oxidized. If we use the violence of the reaction in the two remaining classes as an index of the tendency of the metal to be oxidized to its aqueous ions, then the third class is more readily oxidized than the second. Thus, the grading is third class > second class > first class.
22
Before reading furthel; do Activity 2.2.
In this activity, the second sequence of Reactions of metals on the CD-ROM, you will see the results of adding zinc to a solution of copper sulfate, and copper to a solution of zinc sulfate. Write out an equation for any reaction that occurs. The reddish-brown film formed on the zinc is copper metal, and the simultaneous fading of the blue colour from the solution is consistent with this, because the metal is formed from the Cu2+(aq)ions in the solution. Because zinc forms colourless Zn2+(aq)ions (see Activity 2. l), a reasonable equation is Zn(s) + W + ( a q ) = Zn*+(aq) + Cu(s)
(2.12)
What is oxidized and what is reduced during this reaction? Zinc atoms lose electrons, whereas copper ions gain them. Thus, zinc metal is oxidized and Cu2+(aq)ions are reduced. What happens when copper metal is dropped into a solution of zinc sulfate? Nothing: in other words, the reverse of the reaction in Equation 2.12 does not occur. Copper metal and zinc sulfate solution are the products left when the reaction of zinc metal with copper sulfate solution has ended. The equilibrium position in the system represented by Equation 2.12 lies very far to the right: the favoured combination is zinc in the oxidized form, Zn2+(aq),and copper in the reduced form, Cu(s) (Figure 2.7). This is the combination produced when zinc metal is oxidized at the expense of the ‘de-oxidation’ (reduction) of Cu2+(aq).So of the two metals -zinc and copper -zinc is the more readily oxidized to aqueous ions. Now this is consistent with Section 2.2, where we concluded that as H+(aq) will oxidize zinc but not copper, zinc is the more readily oxidized of the two metals. Let us therefore assume that this consistency with the classification of Section 2.2 is of general application.
Figure 2.7 Reaction 2.12 shows that the combination of copper metal and a solution of zinc sulfate (right) is preferred to, or more stable than, the combination of zinc metal and a solution of copper sulfate (left). 23
In the light of this assumption, predict answers to the following questions. When you have made your predictions, you should do Activity 2.3. What do you expect to happen when (i) magnesium is added to a solution containing Ag+(aq), (ii) copper is added to a solution containing Mg2+(aq), and (iii) magnesium is added to a solution containing Cu2+(aq)?
In this activity, the third sequence of Reactions with metals, on the CD-ROM associated with this Book, you will test the predictions that you have just made by watching the three experiments mentioned above. We now return to the predictions that you were asked to make before you did Activity 2.3. According to the classification of Section 2.2, magnesium is a member of the third class, and silver is a member of the first, so magnesium should be more readily oxidized than silver. The magnesium should be oxidized to Mg2+(aq),and Ag+(aq) should be reduced to silver metal when magnesium is added to silver nitrate solution. Write an equation for the reaction. As the ions are of different charge, conservation of charge can be achieved only by making two silver ions react with each magnesium atom: Mg(s) + 2Ag+(aq) = Mg2+(aq) + 2Ag(s) (2.13) Likewise, because copper is in the first class along with silver, magnesium should also reduce Cu2+(aq)to metallic copper: (2.14) Mg(s) + Cu2+(aq)= Mg2+(aq) + Cu(s) This reaction does indeed occur, but the reverse reaction does not: copper undergoes no reaction when dropped into a solution of magnesium sulfate. We see then that the results of Activities 2.2 and 2.3 are in agreement with the classification made in Section 2.2. Moreover, they suggest an enlarged scope for the classification. Not only do the classes give us an idea of the tendency of a metal to be oxidized by H+(aq), but there seems a possibility that they may also tell us whether or not a metal will reduce the aqueous ions of a metal in a different class.
Now try Questions 2.2 and 2.3.
Describe what will happen, if anything, when pieces of the metals rubidium, strontium and silver are dropped into dilute hydrochloric acid. Write equations for any reaction that you think will occur.
Predict the consequences, if any, of mixing the following chemicals; write equations involving aqueous ions where you think any reaction occurs: (i) copper and an aqueous solution of tin dichloride (SnCl,); (ii) tin and an aqueous solution of silver nitrate; (iii) magnesium and an aqueous solution of tin dichloride; (iv) iron and an aqueous solution of calcium chloride. 24
In this Section, we use an ionic model in which the solid halides and oxides of metallic elements are regarded as aggregates of positive metal ions and halide (X-) or oxide (02-) ions. Consider the formation of two chlorides of this type from the metal and chlorine. Caesium, an alkali metal, forms a chloride CsC1; gold, a noble metal, forms a chloride, AuC13: Cs(s)
+
Au(s)
+ :C12(g)
iC12(g) = CsCl(s) = AuC13(s)
(2.15) (2.16)
In the ionic model, CsCl is regarded as Cs+Cl-, and AuC13 as A u ~ + ( C ~ - ) ~ . What happens to the metal and chlorine in Reactions 2.15 and 2.16? Several things, but the relevant one here is that the metal is oxidized and the chlorine is reduced. The metal atoms lose electrons and become positive ions in the chloride; the chlorine atoms, bound together in diatomic molecules, gain electrons and become C1- ions. As usual, oxidation and reduction are complementary processes. Although both caesium and gold react with chlorine gas, there is an enormous difference in the vigour of the two reactions. At room temperature, the reaction of gold is very slow, but caesium explodes when exposed to a plentiful supply of chlorine. When the two chlorides are gently heated, nothing happens to caesium chloride, but the gold compound decomposes, and the metal and chlorine gas are regenerated: AuCl,(s) = Au(s)
+
5C12(g)
(2.17)
These observations suggest that gold has a much lesser tendency than caesium to be oxidized by chlorine; not only does it combine with the halogen less vigorously, but the metal can easily be regenerated from the chloride. A very similar pattern is observed in the reactions of the two metals with the other halogens, and with oxygen. Caesium reacts explosively with fluorine and oxygen, and combines violently with bromine and iodine when gently heated. By contrast, gold can be melted in air or oxygen without chemical change (Figure 2.8), and combines with bromine and iodine in mild reactions to give halides that, like the trichloride, regenerate the metal on heating. In addition, we know that caesium reacts explosively with dilute acids and water; gold, on the other hand, is unaffected. A chemical that brings about oxidation is called an oxidizing agent; one that brings about reduction is called a reducing agent. We have now considered seven oxidizing agents -the four halogens, oxygen, water and the aqueous hydrogen ion. Whichever oxidizing agent we choose, the reactivity of caesium metal is always greater than that of gold. Suppose now that this result could be generalized to include all metals. It would then be possible to produce a unique grading of metals in order of their tendency to be oxidized, a grading that is valid whether the oxidizing agent is aqueous hydrogen ion, chlorine, bromine, oxygen, or whatever. If this were the case, then the classification that we made in Section 2.2 in terms of the tendency to be oxidized by hydrogen ions would also
Figure 2.8 (a) Caesium reacts instantly with water and air; (b) gold can be melted at over 1 000°C in air, and then poured and cooled without chemical change. 25
apply to the tendency to be oxidized by oxygen, chlorine, etc. By thinking again about what you did in Question 2.3, you should be able to imagine how useful such a grading would be in predicting the consequences of putting one metal in contact with the compound or aqueous ion of another. We shall return to this idea in Section 4, but first we shall see how some support for it can be found in the general chemistry of the extraction of metals from their ores.
1 2
3
4
5
Oxidation is commonly defined as the loss of electrons, and reduction as the gain of electrons. We have tried to classify metals according to the vigour of their reaction with H+(aq). In the first class, we included metals like copper, silver and gold, which undergo no reaction; in the second, we included metals like iron, tin and zinc, which react mildly; in the third class, we included metals like the alkali and alkaline earth metals, which react vigorously.
It then appears that metals in the third class will reduce the aqueous ions of metals in the first and second classes; likewise, metals in the second class reduce the aqueous ions of metals in the first. However, reactions of the reverse type do not happen. Metals in the third class combine more vigorously with oxygen and chlorine than do metals in the first class. Taken together, these observations suggest that it might be possible to arrange metals in a series in order of their tendency to be oxidized. Such a series would have valuable predictive powers. QUESTION 2.4
Use the definitions of oxidation and reduction employed in this Book to determine what element, if any, is oxidized and what element, if any, is reduced in the following reactions: (i)
Sn(s) + Cu2+(aq)= Sn2+(aq) + Cu(s)
(iv)
+ 2Ag+(aq) = Pb2+(aq)+ 2Ag(s) C12(g) + 2Br-(aq) = Br2(aq) + 2CI-(aq) Ba(s) + C12(g) = BaC12(s)
(v)
Ag+(aq) + Cl-(aq) = AgCl(s)
(ii) Pb(s) (iii)
+ 2AgI(s) = Zn12(s) + 2Ag(s) (vii) A1C13(s) + 3K(s) = 3KCl(s) + Al(s) (viii) Fe203(s) + 2Al(s) = 2Fe(s) + AI2O3(s) (ix) 2FeO(s) + O,(g) = Fe203(s) (x) 2Fe2+(aq)+ C12(g) = 2Fe3+(aq) + 2Cl-(aq) (vi) Zn(s)
26
As the behaviour of caesium and gold reveals, a particular metal often responds in a similar way towards quite different oxidizing agents. This observation would be more striking if we were not so familiar with its practical consequences. We can begin with the jewellery shown in Figure 1.2 (p. 14). Like so many metallic objects found on archaeological sites, it is made of gold. Other metals favoured by ancient craftsmen were silver and copper. Gold and silver metals were often found free, but most of the copper was extracted from ores. Now, many of the copper objects discovered at archaeological sites are heavily corroded (Figure 3. l), but those of silver and, especially, gold have suffered much less. What connection exists between the last sentence and the one preceding it? The fact that copper has to be extracted from its ores suggests that it forms compounds more readily than silver or gold. At the same time, this greater willingness of copper to form compounds accounts for the corrosion of the artefacts.
.
Figure 3.1 This Byzantine copper coin from Laodiciae in Turkey reveals obvious signs of copper corrosion.
In other words, the more readily a metal is oxidized, the more difficult it will be to obtain it from its ores. One further example of our ability to grade metals according to their tendency to be oxidized is a qualitative classification in terms of the ease with which we can extract them from their ores. We can distinguish three classes whose composition is very similar to those given in Section 2.2. The so-called ‘noble’ metals, which seem reluctant to react with other substances. They are often found free or can be obtained by heating their ores in air. This first class includes gold, silver and mercury. More easily oxidized metals, which are often obtained by heating their ores in air to form oxides, then heating the oxides with some form of carbon at furnace temperatures up to about 1 500 OC, and then cooling; for example PbO(s) + C(s) = Pb(s) + CO(g) (3.1) This second class includes copper, lead, tin, iron and zinc. The third class of metals has a greater tendency to be oxidized than the first two. It includes the alkali metals and the alkaline earth metals, which, because they cannot readily be obtained by heating their compounds with carbon, were unknown as late as 1800. It was the invention of electrical cells and electrolysis that first enabled chemists to extract these metals from their ores. To reinforce these points, we now consider the extraction and uses of one metal from each of the three classes. The metals we have chosen are mercury, tin and aluminium.
27
Metallic mercury (Figure 3.2) is a heavy silvery liquid with a density nearly 14 times that of water. In nature, mercury is only rarely found in quantity as the free metal; by far the most common source is cinnabar, mercury sulfide, HgS (Figure 3.3). The metal can be obtained by simply heating the ore in air, in a furnace:
(3 .a
H g W + 02(g) = Hg(g) + S02(g> The mercury vapour can then be easily condensed to the liquid metal below its boiling temperature of 357 "C.
Figure 3.2 Mercury is such a dense liquid that (a) cannonballs float in it. (b) Floating human beings seem to barely penetrate the surface.
Figure 3.3 Mercury sulfide, cinnabar, being heated in order to obtain mercury. 28
Figure 3.4 The famous experiment of Antoine Lavoisier: (a) mercury is boiled in contact with a sealed volume of air; (b) the red powder thus formed is heated more strongly.
Science has benefited very greatly from the unusual properties of mercury. Chemists in particular recall the famous experiment (Figure 3.4a) of Antoine Lavoisier ( I 734-1794), which established the nature of combustion and made their subject a modern science. Boiling mercury in the retort was kept in contact with a sealed volume of air in the bell jar and the body of the retort for 12 days. The volume of air decreased by 130 cm3 to about five-sixths of its former value, and a red powder appeared on the surface of the mercury in the retort. Animals placed in the residual air ‘suffocated in a few seconds’, and a burning taper ‘was extinguished as if it had been immersed in water’. Use modern chemistry to explain these facts and to identify the red powder. The boiling mercury has removed nearly all of the 20% of oxygen in the air by reacting with it and forming a mercury oxide -the red powder. After the retort had cooled, Lavoisier scraped all of the red powder off the mercury surface and heated it very strongly (Figure 3.4b). A gas was produced and the red powder disappeared, leaving mercury in its place. The volume of gas was 130 cm3. A taper ‘burned in it with a dazzling splendour, and charcoal, instead of consuming quietly as it does in common air, threw out such a brilliant light that the eyes could scarcely endure it,. If the sample of new gas was added to the lethal gaseous residue from the experiment of Figure 3.4a, the combination possessed all the properties of ordinary air.
So what happens in the experiment of Figure 3.4b? 29
The oxide of mercury decomposes, regenerating the oxygen that had been removed from the air in Figure 3.4a in a pure state, The other product of the decomposition is metallic mercury. For Lavoisier, these experiments revealed the nature of combustion: a reaction of the burning substance with a component of the air, which he called oxygen. But they also have a broader significance. The first experiment involves synthesis: a chemical compound is made by the combination of mercury and oxygen. The second involves analysis: the compound is split up into mercury and oxygen, two substances that were called chemical elements because they cannot be taken apart in the same sort of way. Lavoisier therefore called the red compound mercury oxide, thus building the distinction between elements and compounds, and the fundamental idea of a chemical element, into the nomenclature of chemistry. Other chemical substances were then brought into this system of nomenclature. As a result, although for the modern chemist, Lavoisier's The Elements of Chemistry (1789) may read like an old-fashioned textbook, previous works on the same subject seem unintelligible. The experiments of Figure 3.4 depend on the delicate equilibrium between mercury and oxygen on the one hand, and mercury oxide (HgO) on the other: 2Hg + 02(g) 4 2HgO(s)
(3.3)
At the boiling temperature of mercury (357 "C), equilibrium lies to the right and mercury reacts with oxygen to form HgO. But at 600 "C, equilibrium lies to the left, and there is decomposition into mercury and oxygen (Figure 3.5).
Figure 3.5 (a) Red mercury oxide, HgO, being formed on mercury boiling in air; (b) the same oxide undergoing thermal decomposition at a higher temperature to mercury and oxygen
Is this behaviour of mercury oxide consistent with our classification of mercury as a class 1 metal? Yes; the oxide of mercury yields the metal if it is simply heated. Because mercury is both a liquid and a good electronic conductor at room temperature, it also has many uses in wiring and switching devices. Another major use of mercury is in the manufacture of chlorine and sodium hydroxide. It is, however, very toxic, and has been implicated in some highly publicized cases of environmental pollution. In 1953, methane-producing microbes in the sludge at the bottom of Minamata Bay in Japan, converted waste mercury metal from a factory into the monomethylmercury ion, Hg(CH3)+(aq). In this form, mercury readily penetrates the walls of living cells, and it entered the food chain, ultimately causing brain damage and death among some of those who ate fish caught in the bay. Because of cases like this, the use of mercury has declined in recent years, notably 30
in electrical batteries, and in pesticides and fungicides, where some products have been banned. World production in 1996 was 2 800 tonnes, less than half that in 1986 and only a quarter of the 1971 figure. More than one-third of 1996 production came from the most famous of all cinnabar mines at Almaden in Spain. Figure 3.2 was photographed at this site, which has been in use since Roman times.
The chief ore of tin is cassiterite, Sn02, and this is the only one from which the metal is extracted (Figure 3.6). Currently, the main suppliers of tin ore are China, Indonesia, Peru and Brazil. Much tin is obtained by heating cassiterite with coal in a furnace at 1 200-1 300 OC, when liquid tin and oxides of carbon are formed; for example Sn02(s) + 2C(s) = Sn(1) + 2CO(g) (3 -4)
Figure 3.6 Cassiterite, the principal ore of tin (left), and metallic tin (right).
Since the two reactants are solids, intimate contact is inhibited, and the process would be slow if Equation 3.4 were the sole reaction. However, gas/solid reactions are much faster, and the carbon reacts with carbon dioxide, which is always present in the furnace atmosphere: ~ C ( S+) 2C02(g) = 4CO(g) (3.5) A second gadsolid reaction between carbon monoxide and S n 0 2 then yields liquid tin: 2CO(g) + Sn02(s) = 2C02(g) + Sn(1)
(3-6)
What is the result of adding Equations 3.5 and 3.6? It is Equation 3.4, the overall reaction. Various furnaces can be used, and Figure 3.7 shows the most common type, known as a reverberatory furnace. The burner is fired by oil, gas or pulverized coal, and the flame, with its intensely hot combustion products, is reflected or ‘reverberates’ from the curved roof on to the charge, which is fed in at intervals through pipes in the roof. The charge consists of tin ore, coal and a ‘flux’ such as lime, which promotes 31
Figure 3.7 A reverberatory furnace of the type used to make metallic tin. A typical furnace is about 30 m long and about 10 m wide.
fusion. The latter forms calcium oxide, CaO, in the furnace; in turn this combines with impurities, such as Si02, to yield a liquid slag. The slag floats on the liquid tin, which collects on a hearth at the base of the furnace, where it can be tapped off and solidified. At this stage, the tin contains impurities, such as iron and arsenic. These are much reduced by heating to just above the melting temperature of tin, and drawing the liquid from a solid 'dross', containing impurities with higher melting temperatures. World production of metallic tin in 1997 was 23 1 000 tonnes; 70 000 tonnes of it came from China, 40 000 tonnes from Indonesia, and 40 000 tonnes from Malaysia. The South Crofty, the UK's last operating tin mine closed early in 1998, signalling the end of the great Cornish tin industry of the past. Over one-third of tin production is consumed as tin plate in the canning industry, and about a quarter in solders. The melting temperatures of tin and lead are 232 "C and 327 "C, respectively, but a typical general engineering solder for electronics contains 60% tin and 40% lead, and melts well below either temperature in the range 183-188 "C. Standard plumbers' solders, which contain a higher proportion of lead (about 70%) have a higher melting range (1 83-255 "C).
The oxide of aluminium, A1203,occurs in many parts of the world in combination with between one and three molecules of water. These deposits are called bauxites (Figure 3.8a). In the extraction of aluminium, bauxite is first heated under pressure with 4 mol litre-' NaOH solution at 140 "C. This procedure takes advantage of an unusual property of A1203.It is an oxide that is willing to dissolve in strong bases as well as in acids; such an oxide is said to be amphoteric. The ion Al(OH)4-(aq), in which the coordination is tetrahedral (Figure 3.9) is the main product: A1203(s) + 3H20(1) + 20H-(aq) = 2A1(OH)4-(aq) (3.7) Insoluble impurities, such as the oxides of iron, silicon and titanium, can then be filtered off. When the solution is cooled and diluted, the equilibrium swings back to the left and A1203can be reprecipitated in a hydrated form as the hydroxide, AI(OH)3. If this is heated at 1 100 "C, pure A1203 is formed: 2Al(OH)3(s) = A1203(~)+ 3H2O(g)
32
(3.8)
Figure 3.8 (a) Bauxite, the principal ore of aluminium, is usually red-brown due to the presence of iron minerals; (b) aluminium metal.
Figure 3.9 The Al(OH)4- ion.
The next step is the extraction of the metal from the oxide. Aluminium cannot be obtained by carbon reduction because this yields a carbide of aluminium. Sodium can be obtained by the electrolysis of molten sodium chloride: sodium ions in the melt move to the negative electrode and are deposited as sodium metal. Similarly, if A1203is converted into aluminium chloride, A1Cl3, one might then hope to obtain aluminium in like fashion. Attempted electrolysis, however, is crowned with disappointment. The fused chloride does not conduct electricity, so the melt evidently contains covalent molecules rather than ions. Owing to these difficulties, the first authenticated samples of metallic aluminium were obtained by a non-electrolytic method, employing reduction of the trichloride with metallic potassium: 3K + AlC13 = A1 + 3KC1 (3.9) This type of reaction, using alkali metals, was first operated on an industrial scale by a Frenchman, H. E. St-C. Deville. In 1854, at international exhibitions in Paris, guests of the extrovert Emperor Napoleon I11 enjoyed the privilege of dining with aluminium cutlery. At this time, aluminium cost about &50per pound, because the two reactants in the manufacturing process had to be made from raw materials by rather expensive methods. Moreover, potassium and aluminium trichloride are difficult to handle because they both react with water. Clearly, a new process working directly with the oxide was needed, and the solution was provided almost simultaneously by a Frenchman, Paul Hdroult, and an American, Charles Hall, in 1886. Both men were 22 years old at the time.
Hall (Figure 3.10), became interested in the aluminium problem while he was a student at Oberlin College, Ohio. Immediately after graduation, he borrowed a battery, a crucible and a few other bits of apparatus from his professor, and worked on the extraction problem in his father’s woodshed. As the oxide fuses at such high temperatures, electrolysis of the pure melt was obviously out, but it occurred to Hall that if he could dissolve the oxide in a suitable solvent, it might tKen be decomposed electrolytically, just as CuC12 can be electrolytically decomposed in aqueous soloution. After some difficulties, he found his solvent: it was cryolite, Na3AlF6, a compound that occurs naturally in large deposits in south-west Greenland.
Figure 3.10 Charles Hall ( I 863-1 914). The son of a protestant clergyman, he studied chemistry at Oberlin College, Ohio where his professor spoke of the fortune that awaited the inventor of an economical method of aluminium extraction. Helped by his elder sister Julia, he converted his father’s woodshed into a laboratory, where he successfully operated a miniature version of the modern extraction process in February 1886. Riches soon followed, but not soon enough for his fiancie who became impatient and broke their engagement. Frugal, religious and shy, Hall then devoted himself to the piano and Oberlin College, which received 5 million dollars in his will. 33
When cryolite is melted at about 1 000 "C and aluminium oxide is dissolved in it, electrolysis with carbon electrodes yields molten aluminium of 99.5% purity at the negative electrode: negative electrode: 2A13+(melt)+ 6e- = 2Al(l)
(3.10)
+ 6e-
(3.11)
positive electrode: 302-(melt) =
O,(g)
At the temperature of the furnace, the positive carbon electrodes are burnt away (as carbon monoxide or carbon dioxide gas) by the oxygen formed on their surfaces. Thus, they must constantly be renewed. The cell is kept hot by the electric current through the melt. A typical example is shown in Figures 3.11 and 3.12. This is the process for aluminium extraction which is still used today. In spite of his success, Hall found difficulty in getting financial support, but in 1889 he arrived in Pittsburgh and persuaded some businessmen to found the Pittsburgh Reduction Company, which subsequently became the gigantic Aluminium Company of America (Alcoa). Very large amounts of power are required to extract aluminium electrolytically, and, if a choice has to be made, plants are usually located near sources of cheap hydroelectric power rather than close to bauxite deposits. Jamaica has huge sources of bauxite, but exports it to countries such as Canada, where there are two very large plants at Arvida, Quebec, on the Saguenay River, and Kitimat, British Columbia. For many years, British aluminium production was carried out at two small plants at Kinlochleven and Fort William. In 1970, their combined production was only about 40 000 tonnes. However, in 1968, the British Government provided loans and investment grants to encourage the construction of three large smelters, each with a production capacity of about 100000 tonnes per annum, at Holyhead, Lynemouth and Invergordon. By 1980 British aluminium production had, in consequence, reached a peak of 374 000 tonnes, but during the economic recession of the early 1980s, the Invergordon smelter was closed, and two years later, production was down to 240000 tonnes.
Figure 3.11 Hall electrolytic cell used for aluminium production. A typical plant may contain as many as 1 000 cells, each about 10 m long and 3 m wide.
34
Figure 3.12 An interior view of an aluminium smelter at Portland, Victoria in Australia.
These events were part of a global pattern. During the early 1980s, the aluminium industry was in crisis, and production underwent a significant redistribution. Across the developed world, in the EU, the USA and especially in Japan, aluminium smelters were closed or mothballed, and production was shifted to countries that were closer to bauxite deposits and had lower energy and/or labour costs. Brazil and Australia increased production some fourfold, Canada by 50%, and Norway became the largest European producer. Why were Canada and Norway so favoured? As noted above, cheap hydroelectric power can mitigate the industry’s energy demands; these countries have it. Such changes, and the growth of the Chinese economy, are apparent in Figure 3.13, which shows world aluminium production for 1978-1998 (Figure 3.13a), together with data for some individual regions and countries (Figure 3.13b). In 1998, the world figure was 22.1 million tonnes, British production being 260 000 tonnes. Currently, canning and, especially, transport (Figure 3.14) are potential growth areas for the aluminium industry. With a density of a little more than a third that of stainless steel, aluminium has been associated with the aerospace industry from the very beginnings of powered flight, the engine casing of the Wright brothers’ aircraft being made of it. Even today, aluminium remains the dominant metal in all but the most high-performance aircraft.
3s
Figure 3.13 (a) Recent world aluminium production figures; (b) production data for some individual countries.
Figure 3.14 (a) The Japanese bullet trains with a top speed of 285 km h-' have a double skin of aluminium. (b) Aluminium is the principal structural metal in Jumbo jets; the Boeing 747 contains 75 tonnes.
Our division of metals into three classes according to their tendency to be oxidized correlates well with the relative difficulty of extracting metals from their ores. For example, mercury (in class 1) occasionally occurs naturally, and can also be obtained merely by heating cinnabar, HgS, in air. Tin (in class 2) is obtained by roasting cassiterite, Sn02, with carbon. Aluminium (in class 3) requires much energy for its extraction, which involves the electrolysis of the pure oxide in molten cryolite (Na3A1F6)using graphite electrodes.
36
In Sections 2.2-2.4 we examined a series of experiments and observations which suggest that it might be possible to produce a single grading of metals arranged in order of their tendency to be oxidized by various oxidizing agents. This idea is not a new one; it was first propounded in the nineteenth century, and several chemists advanced quite detailed gradings, which were largely based on a more thorough series of qualitative observations than ours. These gradings are often called the activity series of metals. A typical example from a textbook of that time is given below: Cs > Rb > K > Na > Li > Ba > Sr > Ca > Mg >Al>Zn>Fe>Sn>Pb>H>Cu>Hg>Ag>Au Hydrogen has also been included in the series because many metals reduce hydrogen ions to hydrogen gas. Is this series consistent with the classifications advanced in Sections 2.2 and 3? If so, where do the dividing lines between the classes come in the series? It is consistent. The dividing lines come between aluminium and zinc, and between hydrogen and copper. If this single series correctly expresses the relative tendency of metals to be oxidized by any oxidizing agent, it could be used to predict the outcome of many metal reactions. Whereas before we assigned metals to one of three classes on the basis of their tendency to be oxidized, we now have a series that grades the metals within the classes. Test whether you have properly appreciated this by doing Question 4. I now.
By assuming that the classification or grading of metals made in Sections 2-4 correctly represents ‘the tendency of metals to be oxidized’ by a variety of oxidizing agents, predict the outcome, if any, of an attempted reaction in which the following pairs of substances are heated together. Write balanced equations if you think reactions will occur. (In most cases we have not included physical states because they will vary according to the temperature of the reaction.) Cu and Zn12 (i) (ii) Zn and CuO (iii) Zn and FeC12 (iv) Fe and KI (v) K and A1C13 (vi) Hg andA1203 (vii) Zn(s) and Ag+(aq) (viii) Fe and PbBr2 (ix) Zn(s) and H20(g) (x) Ba and FeBr2
37
You should now do Activity 4.1.
In this activity, the fourth sequence of Reactions of metals, on the CD-ROM associated with this Book, you will see metals added to solutions of the salts of other metals. You are asked to decide whether a reaction occurs. The reactions are performed first in the cold and then, in some cases, warmed to speed up any reaction. As most of the solutions are colourless, you will need to make especially careful observations of any changes in the nature of the metal. Blackening or swelling are frequent signs of reaction when a new metal is precipitated or formed on the surface of another. The following reactions are attempted: (i) (ii) (iii) (iv)
iron filings and copper sulfate solution; copper and silver nitrate solution; magnesium and zinc sulfate solution; magnesium and lead nitrate solution;
aluminium and copper sulfate solution; (vi) aluminium and dilute hydrochloric acid; (vii) zinc and lead nitrate solution. You will find that some metals cause evolution of hydrogen gas from the solution, especially on warming. When you have finished, consider whether your observations are consistent with the implications of the series given at the beginning of this Section. Explain why you think they are consistent or not consistent as the case may be. For a comment on the experimental results, see p. 40. (v)
Before beginning the next Section, attempt Question 4.2.
The dissociation of the weak acid, acetic acid, in aqueous solution, may be written HAc(aq) = H+(aq) + Ac-(aq) Write down an expression for the equilibrium constant of the reaction.
It is now time to examine critically our proposal that a single grading of metals can correctly express their tendency to be oxidized by various oxidizing agents. First then, let us turn to the language that we have used. Words and phrases such as ‘tendency to be oxidized’, ‘willingness to be oxidized’, ‘violence of reaction’ and ‘reactivity’ do not describe quantitative concepts that can be precisely defined and expressed in measurable units. This is a serious limitation, which hinders extension of the ideas if they are correct, and prevents the uncovering of their weaknesses if they are not. Until now, our observations have not been quantitative.
38
If you think carefully about some of the observations that we have made, you may already be able to spot weaknesses in the arguments. Consider, for example, the classification in Section 2.2, which was built on observations made in that Section and supported by others in Section 2.3. In Section 2.3 we used the fact that the equilibrium position in the reaction Zn(s) + Cu2+(aq)= Zn2+(aq)+ Cu(s) (2.12) lay to the right-hand side as an indication that zinc was more readily oxidized than copper. What criterion did we use in Section 2.2 to show that magnesium was more readily oxidized than zinc or tin by aqueous hydrogen ions? We used the vigour, or to be more precise, the rate of the reaction. Now you know the importance of distinguishing the concept of how far a reaction goes from how fast it goes, that is, of distinguishing the equilibrium constant from the rate of reaction. The two are often completely unconnected; for example, a reaction like the combustion of natural gas, which has a huge equilibrium constant, proceeds immeasurably slowly at room temperature (in the absence of a spark!), Yet now we are trying to grade metals according to their tendency to be oxidized, by considering at the same time both rates and positions of equilibrium. The idea of grading the metals according to qualitative observations makes an immediate appeal to the senses, but it may be that we have allowed the visual force of the idea to bring us into conflict with important chemical principles. What we badly need, is an approach in which the reactions of metals are compared quantitatively under the same conditions in terms of either rates or positions of equilibrium. Which of these two approaches is likely to be the more helpful? We can answer this by noting that, in developing our ideas, we have discussed several reactions that do not occur: reactions such as Cu(s) + 2H+(aq) = Cu2+(aq)+ H2(g) (2.9) Cu(s) + Mg2+(aq)= Cu2+(aq)+ Mg(s) (4.2) Clearly, we cannot determine the rates of these reactions if we cannot detect their occurrence. This suggests that an approach in which we compare the reactions of metals by their rates would be quite unproductive in certain important cases. This leaves us with the approach in which we compare equilibrium constants. The expression that we wrote for an equilibrium constant, K, in Question 4.2 suggests that values of K can be determined by finding the concentrations of the reactants and products in the equilibrium state. However, Reaction 4.2 occurs to such a small extent that it is not possible to determine the concentration of Cu2+(aq)-written [Cu2+(aq)]-by orthodox chemical analysis. You may therefore raise the same objection to the comparison of equilibrium constants that we raised to the comparison of rates: how can we compare the equilibrium constants of reactions that do not appear to occur? As we shall show you between this point and the end of Section 11, the science of chemical thermodynamics enables us to do just that. For this reason, we now embark on an approach in which we compare, under the same conditions, the equilibrium positions of the reactions of metals.
39
1
2
40
The methods and observations that we have used to arrange metals in order of their tendency to be oxidized are not quantitative; nor do they make the necessary distinction between the equilibrium constant and the rate of a chemical reaction. A comparison of metal reactivity based on equilibrium constants is likely to be more productive than one based on reaction rates.
Let us start by trying to formulate the problem in a more precise way. One way of recognizing equilibrium in a chemical system is by the ‘appearance of quiescence’ : nothing changes as time passes. Once at equilibrium, or in an equilibrium state, a chemical system does not change with time unless disturbed in some way: thereafter it heads inexorably back toward equilibrium. With this in mind, consider ordinary table salt, sodium chloride. This is a compound of sodium and chlorine. However, under normal conditions, the table salt you use at home does not decompose into sodium metal and chlorine gas of its own accord. (Perish the thought!)
To be more precise, at 298.15 K (25 “C) the equilibrium position for the reaction NaCl(s) = Na(s)
+
:Cl,(g)
(5.1)
evidently lies so far over to the left-hand side that it effectively does not happen at all. In other words, solid sodium chloride (rather than a mixture of sodium and chlorine) corresponds to the equilibrium state for this chemical system. Further evidence is provided by the experimental observation that the reverse of Equation 5.1 , the reaction between metallic sodium and chlorine, Na(s)
+
;-Cl,(g) = NaCl(s)
( 5 -2)
happens spontaneously: the equilibrium state again corresponds to NaCl.
So far so good. Seen in a broader context, however, the chemical system described above shares a feature in common with other naturally occurring (or spontaneous) processes: it has a direction. For example, the spontaneous reaction in Equation 5.2 behaves rather like a stone rolling downhill: there is a ‘downhill’ character to this reaction that corresponds to the movement toward equilibrium.
Figure 5.1 Inanimate objects do not roll uphill!
By contrast, there is an ‘uphill’ character to its reverse -Equation 5.1. Just as balls, etc., don’t roll uphill on their own (Figure 5.1), so too the movement away from equilibrium in this chemical system can be achieved only with the help of outside assistance, for example, by melting the sodium chloride and electrolysing it. These ideas apply equally well to all chemical reactions. Any spontaneous reaction moves toward, never away from, equilibrium. But what is the underlying reasun for this behaviour? What determines the equilibrium state in a chemical system, and hence governs whether a reaction can happen or not?
41
One way towards an answer to the question posed at the end of Section 5 is suggested by the analogy we used earlier: when a stone rolls downhill, we explain its motion by saying that it happens because the stone thereby achieves a state of lower gravitational (or potential) energy. Is there a model here for chemical reactions? Perhaps a chemical reaction can occur only if the products are of lower energy than the reactants? Reactions are classified as either exothermic (releasing energy to the surroundings as heat) or endothermic (absorbing energy from the surroundings as heat). The immediate effect of an exothermic reaction is generally a rise in the temperature of the reaction mixture. For instance, when aluminium is dropped into a dish of liquid bromine, a reaction occurs and the dish gets hot (Figure 6.1). But common laboratory glassware, like a beaker or test-tube, is a rather poor heat insulator: as soon as the temperature rises above that of the surroundings, heat begins to escape from the reaction vessel. This simple observation points to the one universal statement that can be made about exothermic reactions: if the products of such a reaction arefinally obtained at the same temperature as the reactants, then heat must have left the reaction vessel. This, and the complementary case of an endothermic reaction, both at constant temperature*, are illustrated schematically in Figure 6.2. With the additional proviso that the reaction also takes place at constant pressure?, the heat (4) released or absorbed by a reaction at constant temperature can then be identified with the enthalpy (from the Greek en, meaning ‘in,, and thalpi!, meaning ‘make hot’) of reaction, AH, at that temperature. Thus,
AH
= q (at constant T and p )
(6’
Remember that the symbol A (Greek capital ‘delta’) means a change in some physical quantity, in this case a change in enthalpy, H . So, for a reaction at constant T and p , AH = H(products) - H(reactants) (6.2) To summarize: Equations 6.1 and 6.2 together imply that the heat released (or absorbed) by a reaction is a measure of the difference in energy between reactants and products under the same conditions of temperature and pressure.
* Throughout this Book, the phrase ‘at constant temperature’ simply implies that the initial and final temperatures are the same (as shown in Figure 6.2), no matter what happens in between. It does not necessarily imply that the temperature is actually held constant throughour the reaction, as in a thermostat for example, although this may indeed be so in certain cases. 7 This condition derives from the fact that laboratory reactions are usually carried out in vessels open to the atmosphere, that is, at constant (atmospheric) pressure. For this reason, chemists chose to define a property, enthalpy, which can be used to express the energy transferred under this condition. A further comment on conditions governing the validity of Equation 6.1 is discussed in the Appendix. 42
Figure 6.1 (a) Aluminium foil immediately after addition to liquid bromine. (b) A violent exothermic reaction has begun: much heat is evolved, along with fumzs and light.
Figure 6.2 Schematic representation of (a) an exothermic reaction, and (b) an endothermic reaction, taking place at constant temperature.
What are the signs (i.e. positive or negative) of AH for an exothermic reaction and for an endothermic reaction? This derives from the sign convention for heat transfer: q is positive (> 0) if heat is absorbed by the system of interest (the reaction mixture in this case), but negative (< 0) if heat is released to the surroundings. From Equation 6.1, therefore, AH is negative for an exothermic reaction and positive for an endothermic one. In the context of Equation 6.2 therefore, only exothermic reactions result in products of lower energy than the reactants. (This is shown schematically in Figure 6.3a.) The energy analogy with the motion of a stone, proposed at the start of this Section, therefore implies that only exothermic reactions can happen spontaneously, whereas endothermic reactions (for which the products are of higher energy than the reactants, Figure 6.3b) cannot occur on their own. In 1853, this hypothesis was proposed by the Danish chemist, Julius Thomsen (Figure 6.4).
Figure 6.3 Schematic representation of (a) an exothermic reaction, and (b) an endothermic reaction, defined in terms of enthalpy changes at constant temperature and pressure.
You should now try the following Question, to see how this hypothesis compares with the observed behaviour of a chemical system.
Figure 6.4 Julius Thomsen (1 826-1 909). This photograph captures the solitary nature of this Danish chemist, who determined many important enthalpies of reaction, while refusing frequent offers of collaboration from others. This partly explains the accuracy of his values, because foreign hands were never able to tamper with the internal consistency of his work. He also published (1 895) an early long form of the Periodic Table (subsequently improved by Alfred Werner), which was used by Niels Bohr, his fellow countryman, to deduce the electronic configurations of atoms. For this, Bohr won the Nobel Prize for Physics in 1922.
43
Consider the reaction between sodium and chlorine discussed in Section 5 : Na(s)
+
1
TCl,(g) = NaCl(s)
(5.2) If one mole of sodium reacts with half a mole of chlorine at 298.15 K (25 OC), the enthalpy change is:
AH = -41 1.2 kJ (a) Is the reaction exothermic or endothermic? What is the value of AH for the reverse reaction, Equation 5. I? (b) Could Thomsen’s hypothesis provide a satisfactory explanation for the observed behaviour of this system, as outlined in Section 5? Seen in a more critical light, however, just one example of a spontaneous endothermic process would be sufficient to discredit Thomsen’s hypothesis. At room temperature, it is quite difficult to find such examples, but not impossible. One mole of table salt will dissolve spontaneously in a large amount of water: NaCl(s) = Na+(aq) + Cl-(aq); AH = +3.9 kJ at 298.15 K (6.3) Similarly, the use of thionyl chloride (S0Cl2)as a common dehydrating agent in the laboratory depends on another equally spontaneous reaction; when one mole of each reactant is involved, we have: S0Cl2(1) + H20(1) = SO&) + 2HCl(g); AH = +47.1 kJ at 298.15 K (6.4) Evidently, chemical reactions do not ‘roll downhill’ on a simple ‘enthalpy landscape’. Nevertheless, as the example in Question 6.1 suggests, there is some merit to Thomsen’s hypothesis. Under ambient conditions, the great majority of spontaneous reactions are exothermic. Because of this, it seems reasonable to suggest that the enthalpy change for a reaction is one factor that influences the equilibrium state. But, given the examples cited above, we shall expect to find that it is not the only one.
1
2
44
The enthalpy change, AH, is negative for an exothermic reaction and positive for an endothermic one. At normal temperatures, most spontaneous reactions are exothermic (negative AH). This raises the prospect of some form of energy change that is always negative for a spontaneous reaction. In contrast to the predictions of Thomsen’s hypothesis, AH will be an important, but not the only, contributor to this energy change.
The discussion in the previous Section brings us back to our original question, but with a little more insight into the sort of answer to expect. As we said in Section 5 , the problem is one of predicting the direction of spontaneous change in a chemical system. Questions like this lie within the province of one of the most important and far-reaching laws in the whole of science, You may have heard of it: it is called the second law of thermodynamics.As you will see, it is this law that provides a criterion for equilibrium in a chemical reaction. Perhaps the most commonly expressed form of the second law of thermodynamics is the one proposed by the German scientist Rudolf Clausius in 1865: ‘the entropy of the universe tends towards a maximum’. Unfortunately, Clausius’s famous aphorism is a little too unrestrained to be useful, so we shall adopt the following statement:
This means that if, in the world around us, something happens in a region which, during the time-scale of the event, we can regard as an ‘isolated system’, then, AStotal
’0
(7.1)
where A again means ‘a change’, and S denotes the entropy (from the Greek en, meaning ‘in’ and trope meaning ‘transformation’). We shall treat the second law as one of the ‘rules of the game’ of thermodynamics -that is, as a basic assumption that can be vindicated only by experiment. Our faith is not misplaced! Deductions from the second law have never yet been shown to be invalid, and this success has impressed greater minds than ours (Figure 7.1). Before going further, however, notice that the above formulation raises two immediate questions: What is an isolated system? What is entropy?
To deal with the first: the word system simply means the collection of materials, chemicals, etc., that we are interested in. Once the system is defined, everything else is the surroundings. An isolated system is then one that can exchange neither energy (especially heat) nor matter with its surroundings. For example, a reaction mixture enclosed in a sealed, perfectly insulated container (a sort of ideal vacuum flask) would be an isolated system. Are these the conditions under which most chemical reactions are carried out?
No; most laboratory reactions are carried out in poorly insulated containers, often open to the atmosphere. Under these conditions, reactions readily exchange heat (determined by the sign and magnitude of AH) with their surroundings.
Figure 7.1 Albert Einstein (1 879-1 955). A loose translation of one of his many quotable rernarks runs as follows. ‘The most impressive theories are those which combine very simple premises with a very broad range of applications, and which establish relationships between many things that might otherwise stay unconnected.. .. I am convinced that thermodynamics is the only universal physical theory that will never be overthrown. ’ 45
Thus, reacting chemicals, by themselves, cannot normally be regarded as an isolated system. So as you will see, to apply the second law to a chemical reaction carried out under normal laboratory conditions, our definition of the system must extend beyond the boundaries of the chemicals. But what about the second, more fundamental question: what is entropy?
The question that we have just posed can be approached in two very different ways. Historically, the concept of entropy emerged as a natural development of thermodynamics. Despite its central importance as the arbiter of the direction of spontaneous change, from this point of view, entropy is just another physical quantity -like volume or mass, for example. In other words, entropy can be considered as a property of matter in bulk-just like the more familiar properties mentioned above. The question ‘What is entropy?’ then resolves itself into the more practical question ‘How is entropy measured?’ In this sense, the concept of entropy would be meaningful even if we knew nothing about atoms or molecules. On the other hand, it is possible to start from a theoretical model of how atoms and molecules behave, and hence build up a ‘molecular’ picture of entropy and entropy changes. Unfortunately, the connection between entropy and molecular properties is not at all obvious. However, it is in this context that you may have heard entropy associated with ‘chaos’ or ‘disorder’ or ‘mixed-upness’ or words of a similar sort. This can be misleading, because words like disorder have a rather special meaning in this context -a meaning that is not obviously connected with their everyday sense. One simple example must suffice. At 298.15 K and 1 atm pressure, Avogadro’s law tells us that equal volumes of gases, such as the noble gases helium and xenon, contain the same number of molecules (Figure 7.2). The noble gases are monatomic; that is, the molecules are atoms. It is found experimentally (see Section 10.3) that the entropy of the xenon gas sample is roughly 50% greater than that of the helium. But there is no obvious sense in which the atoms in the xenon sample are the more disordered or ‘disarranged’.
Figure 7.2 Equal volumes of helium and xenon at the same temperature and pressure contain the same number of molecules -in this illustration, eleven! The xenon molecules (left) are larger and more massive than those of helium (right 1, but there is no immediately Iobvious sense in which the xenon sarnple is more disordered.
46
From now on, we shall restrict ourselves to the ‘thermodynamic’ approach and ask how entropy is measured. For this purpose, we ask you to accept the following, essentially operational, definitions of the entropy change in three very important cases (all at constant temperature and pressure): When heat is transferred to or from a pure substance at a temperature T, the change in the entropy of the substance is
where, as usual, the value of 9 is positive if heat is absorbed by the substance. When heat is transferred to or from a large heat reservoir at a temperature T, such as the atmosphere around us, the change in the entropy of the reservoir is given by
When a chemical reaction takes place, the change in entropy during the reaction is given by
AS,,,,,
= S(products) - S(reactants)
(7.4)
According to the definitions in Equations 7.2 and 7.3, what is the SI unit of entropy? The SI units of heat and temperature are the joule (J) and the kelvin (K), respectively. So entropy has the unit joule per kelvin (J K-I). These three cases have been chosen carefully. Case 1 is, as you will see in Section 10.3, used to determine entropies of pure substances. When these have been obtained, case 3 can be used to determine entropy changes in chemical reactions. Finally, case 2 will enable us to discover a new function that does the job we require; that is, it determines the direction of spontaneous change in a chemical system at constant temperature and pressure.
To gain some familiarity with the second law, and with case 1 (Equation 7.2), let us see if they are consistent with an everyday observation. Suppose you have two large blocks of iron -block 1 and block 2, say -at temperatures T I and T2,respectively. Suppose further that T2 > T I ,and that you then touch the blocks together briefly so that a small amount of heat passes from one to the other. Finally, assume that the heat transferred is small enough to leave the two temperatures essentially unchanged. In which direction does heat flow? From block 2 to block 1 -from the block at the higher temperature (T2)to the block at the lower temperature ( T I ) . Suppose that 4 is the amount of heat transferred in this way (Figure 7.3).
Figure 7.3 Two large metal blocks at temperatures T I and T2 (T2> T I )are briefly allowed to touch, or connected by a thin wire, so that a small amount of heat, q, passes from the block at T2 to the block at T I . 47
What are the values of AS for block 1 and block 2, respectively? Block 1 absorbs heat, q, so according to Equation 7.2, AS, = q /T 1 Block . 2 loses an identical amount of heat, so AS2 = -q/T2. If the two blocks are considered as one system, what is the total entropy change?
Taking the two blocks as an isolated system (that is, assuming no heat loss to the surroundings), is the heat transfer in accord with the second law? Yes; because T2 is greater than T1,q/T2 is less than q/T1,so AStOt,1is positive.
Repeat this treatment by considering the case in which heat passes from block 1 at the lower temperature T I to block 2 at the higher temperature T2.Show that this process violates the second law of thermodynamics. In summary: our formulation of the second law, and our definition of entropy change in case 1, leads to the idea that heat cannot pass spontaneously from a colder to a hotter body, which is in accord with everyday experience. Can you see why this must be so? According to the discussion above, transfer of heat from a colder to a hotter body would be accompanied by a forbidden decrease in entropy. Indeed, this statement is an alternative formulation of the second law. Examples like those discussed in this Section reveal the difference between entropy and energy.
Let us now see if the second law can provide the condition that allows a chemical reaction to occur.
As an example of chemical change, consider again the reaction between sodium and chlorine:
Na(s)
+
1
?C12(g) = NaCl(s)
(5.2)
Let us suppose that this reaction takes place at a constant pressure of 1 atm in a large laboratory. Suppose further that we start with one mole of sodium and half a mole of chlorine at a fixed temperature of 298.15 K (25 "C), and eventually finish up with one mole of sodium chloride at the same temperature. 48
Now, when the reaction takes place, there will be an associated change in entropy, ASreac,,given by case 3 (Equation 7.4). As we hinted above, this can be determined from experimental measurements (discussed in Section 10.3): at 298.15 K and 1 atm pressure, it has the following value for the change that we have just described (Reaction 5.2): ASreac. = -90.7 J K-' At first sight, this value seems to be in violation of the second law: the reaction happens, but the entropy decreases! This apparent failure is a reminder of the warning we sounded at the beginning of this Section. Under normal conditions, the reaction mixture is not an isolated system: it affects the surroundings to an extent determined by the sign and magnitude of the associated enthalpy change. Now, the value of AH for this reaction was given in Question 6.1 : AH = -411.2 kJ
So the reaction is exothermic: it releases heat. It follows that the immediate surroundings (the laboratory) have no option but to absorb heat, and hence (according to Equation 7.3) to increase in entropy. Suppose now that, on the time-scale of the reaction, we define our system as the laboratory, and regard it as isolated: no net heat is transferred to it or withdrawn from it by the outside world. According to the discussion above, there are then two contributions to the total entropy change in this system when the reaction occurs: the change due to the reaction (-90.7 J K-l), and the change in the entropy of the Surrounding laboratory, ASlab. We can obtain an expression for ASlab.by using an equation from Section 6. At constant T and p , mreac.
=4
Now, the heat evolved by the chemicals is absorbed by the surrounding laboratory, so Thus,
Notice that because Reaction 5.2 is exothermic, AHreac. is negative. Thus, according to Equation 7.7, ASlab. must be positive, which agrees with the fact that the laboratory absorbs heat. What is the total entropy change for the reacting chemicals plus the surrounding laboratory?
Now according to the second law of thermodynamics, the reaction can occur only if AStotal> 0. According to Equation 7.8, at constant T and p , this condition is:
This result is illustrated schematically in Figure 7.4.
49
Figure 7.4 The components of the total entropy change for the reaction vessel and its surroundings when a spontaneous exothermic reaction occurs at constant temperature and
pressure.
Check that the inequality 7.9 holds for the reaction discussed in this Section: Na(s)
+
Cl,(g) = NaCl(s)
(5.2)
Remember that AH (= -41 1.2 kJ) is given in kilojoules, that AS (= -90.7 J K-l) is given in joules per kelvin, and that T = 298.15 K. Is the second law vindicated? The inequality 7.9 is usually rearranged slightly, by moving the two contributions to the right-hand side and multiplying through by T. If we then drop the subscript ‘reac.’, this gives 0 > AH - TAS; or, rearranging, AH - TAS < 0 (7.10) Thus, the quantity (AH- TAS) does for chemical reactions (at constant T and p ) what potential energy does for the motion of a ball. As we anticipated in Section 6, AH contributes to it, but there is also a second contribution, -TAS. For a reaction to occur spontaneously, the quantity (AH- TAS) must be negative. Conversely, a positive value for (AH - TAS) corresponds to a movement away from equilibrium: such a process never happens of its own accord. All that remains, then, is to find out how to determine AH and AS for a given process at a specified temperature: we can then use the inequality above to predict if a reaction can happen. We deal first with the more familiar quantity, AH.
50
1
The direction of spontaneous change is prescribed by the second law of thermodynamics :
2
For our purposes, the most important point about an isolated system is that it cannot exchange heat with its surroundings. Thus, reactions at constant temperature and pressure are not isolated systems. A change in the entropy of a substance (at constant T and p ) can be defined in terms of heat transfer, q, as AS = q/T. When a reaction occurs at constant T and p , there are two contributions to AStotal:the entropy change associated with the reaction itself, AS = S(products) - S(reactants) (7.4) and the entropy change of the surroundings, -AH/T. According to the second law, the sum of these entropy changes must be positive: this leads to the central inequality for a chemical change at constant T and p , namely: AH - TAS < 0.
3 4
5
At 298.15 K and 1 atm pressure, the values of AH and AS for the reaction of one mole of thionyl chloride with water SOCl,(l) + H20(1) = S02(g) + 2HCl(g) (6.4) are as follows: AH = +47.1 kT AS = +334.6 J K-I Calculate: (i) the entropy change of the surroundings; (ii) the total entropy change, Astotal.Is the sign of AStota1in accord with the observation that this reaction happens under the conditions cited above? What is the necessary requirement for a spontaneous endothermic reaction?
51
In Section 6, we identified AH for a reaction with the heat released or absorbed in order to maintain a constant temperature. But so far we have not considered how to measure heat, nor even given a very precise definition of it. To redeem this failing requires another 'rule of the game' -the first law of thermodynamics.You will be familiar with this law under the guise of 'the law of conservation of energy'. Our first step will be to formulate this principle in a slightly different way, and then to show that it provides a route to the measurement of enthalpy changes. We shall start with a simple and familiar example.
Suppose that you have a beaker of water (100 g, say) at room temperature and pressure, and you want to raise its temperature -to bring about the change shown schematically in Figure 8.1.
Figure 8.1 Raising the temperature of a sample of water from 298.15 K (25 " C )to 323.15 K (50 "C).
The obvious thing to do is to heat the water, and you could do this in a variety of ways. For example, you could stand the beaker on an electric hotplate, or over a flame, or even in front of an electric fire. Or you could simply immerse it in a bath of hotter water. All these methods depend on there being a temperature difSerence between the system of interest (your beaker of water) and the surroundings (the hotplate, flame, etc.). This, in turn, suggests a more precise definition of heat: it is energy transferred because, and solely because, of a temperature difference between a system and its surroundings.
Now think of a different way of bringing about the change in Figure 8.1 -a way that does not depend on a temperature difference. You may have come up with several possibilities, but the simplest alternative is the one you use every time you switch on an electric kettle or an immersion heater. 52
At first sight this may not seem too different from our original suggestion -of using a hotplate, for example. After all, an electrical heating element is involved in both cases and the temperature of the water rises because of the higher temperature of the heating element with which the water is in contact. The crucial distinction is illustrated in Figure 8.2: as you can see, it arises from defining the system so as to include the heating coil. Energy transfer does not, therefore, depend on a temperature difference between the system and its surroundings.
Figure 8.2 Diagram to illustrate the difference between heating a system (b) and doing electrical work on it (a).
The general name for this second sort of energy transfer is work, denoted by the symbol w (electrical work, wel, in the example cited above*). This is the only type of work that we shall consider here. Thus, it seems that we can transfer energy to the water in the beaker either by heating it directly, or by doing electrical work on it; or we could do both together. Whatever method we choose, there will be an increase in the energy of the system, corresponding to heat and/or electrical work that has caused the change. If the change takes place at constant pressure (as it would in a beaker open to the atmosphere), we again identify this increase in the energy of the system with a change in the enthalpy. This leads to the following general expression:
Thus, wel is defined as positive if electrical work is done on the system, and energy is thereby transferred to it. Equation 8.1 is our reformulation of the principle of energy conservation, in terms of the enthalpy of the system. As such, it is a special and strictly restricted case of the first law of thermodynamics-restricted in that it applies only to changes at constant pressure, where the only type of work considered explicitly is electrical * At first sight the name ‘work’ may seem a trifle bizarre in this context: it’s not easy to envisage electricity ‘doing work’ ! The term is a legacy from the early days of thermodynamics, and its application to steam engines and the like. In that context, interest centres on the amount of useful (generally mechanical) work that can be achieved for a given input of heat. As the subject developed, the definition of work was broadened to incorporate all types of energy transfer other than heat.
53
work. Although this condition is satisfied by the examples that you will meet in this and other Books in this series, you should always bear in mind that Equation 8.1 is restricted in this way*. So far, so good; but how does Equation 8.1 help us to measure enthalpy changes? In particular, how do we determine a value of AH for the change implied by Figure 8.1, namely
AH
= H(T2, p ) - H(T1, p ) = ?
(8.2)
where H( T I ,p ) and H( T2,p ) are enthalpies of our sample of water at the initial and final temperatures (298.15 K and 323.15 K), respectively, and at constant pressure p . For the purpose of the argument, suppose that you decide to heat your sample directly by standing it on an electric hotplate. Suppose further that, like most electrical appliances, the hotplate is marked with its power-rating. What is the definition of power? What is its SI unit? Power is the rate of energy conversion: its SI unit is the watt (W), which is defined as follows: W = J s-*. Thus, knowing the power-rating of the hotplate and how long it is switched on, it is a simple matter to calculate the amount of electrical energy converted (= power x time). But there is a snag-a serious one: there is no simple way of telling what proportion of this energy is actually transferred to the system, rather than to the immediate surroundings. Look back at Equation 8.1. In the case under consideration, no electrical work is done on the system (the heating element is not included in the system), so wel = 0. The crucial problem is that we cannot measure the heat q transferred to the system, so we cannot determine the corresponding enthalpy change AH (Equation 8.2) in this way.
Given that AH for a reaction is also defined in terms of heat transfer (Equation 6. l), this is a rather disturbing conclusion. But there is a way around the problem. Our imaginary experiment with a beaker corresponds to one extreme case (and evidently not a particularly useful one) of the general expression in Equation 8.1. What about the other extreme, when q = O? How can this be achieved in practice? The simplest technique is to thermally insulate -for example, by replacing the beaker in Figure 8.1 with a vacuum flask. Under these conditions, the only way to transfer energy to the system is to do electrical work on it. The question that now arises is: does this alternative provide a way to evaluate the enthalpy change in Equation 8.2? To examine this issue, try the following Question.
* To lift this restriction, and hence give a more general statement of the first law, requires the introduction of another, and more fundamental, thermodynamic quantity -the internal energy, U , which we shall ignore. For futher comment on the restricted nature of Equation 8.1, see the Appendix, p. 203. 54
An electrical heating coil with a power-rating of 100W is immersed in 100 g of water in a Dewar flask (the scientific equivalent of a vacuum flask). When the current is switched on for 1 minute and 44 seconds, the temperature of the water rises from 298.15 K to 323.15 K. What is the value of the enthalpy change in Equation 8.2? What assumptions must you make?
So the answer to our question is yes, with the proviso that the system is perfectly insulated. Although an ordinary vacuum flask is far from perfect in this sense, more sophisticated equipment can come close to the ideal limit of no heat transfer. As you saw in Question 8.1, the required value of AH can then be determined by simply measuring the electrical work needed to produce the desired temperature rise. In symbols, = H(T2, p ) - H(T1, p ) = we1
(8.3)
since q = 0.
A perfectly insulated system like this is described as being adiabatically (from the Greek adiabatos, meaning ‘impassable’) enclosed. We conclude with two further points about the first law, which, for the special cases considered here, takes the form:
AH = q + We]
(8.1)
First, notice that Equation 8.1 relates explicitly to a change in enthalpy. The implication is that only diferences in enthalpy can be determined, not absolute values. The problem is analogous to that of measuring the heights of mountains and the depths of oceans. In each case, we can, in fact, measure only a diference in height. Because sea-level is universally accepted as the reference point for measuring height, however, the reference to ‘above’ or ‘below’ sea-level is generally dropped. You will see later how we choose an arbitrary zero for enthalpy. The second point is that the change in enthalpy of a system depends only on its initial andfinal states, and not at all on how it got from one to the other. The ‘state of a system’ simply implies a specification of all the properties necessary to define its condition. For our purposes, listing the temperature, pressure and composition of the system will usually suffice. Thus, in Question 8.1, for example, the initial state of our sample of water (Figure 8.1) could be defined as follows (since the molar mass of water is 18.02 g mol-l): (100/18.02) mol of pure water at a temperature of 298.15 K and a pressure of 1 atm. Likewise, the final state consisted of (loo/ 18.02) mol of pure water at a temperature of 323.15 K and a pressure of 1 atm. In Question 8.1, the change from the initial to the final state was accomplished by putting the sample in an adiabatic enclosure, and performing electrical work on it ( q = 0; AH = we,). However, the same result could have been achieved simply by transferring heat to the system with a hotplate (wel = 0; AH = q). These two methods are quite different, but because in each case, water is taken from the same initial state to the same final state, AH is the same despite the two different pathways. Because of this indifference to the path of change, enthalpy is called a state function.
55
We shall consider two representative examples of enthalpy determinations for reactions -one an endothermic process and the other an exothermic process. In both cases we shall take our cue from the discussion in the previous Section, and concentrate on systems that are, at least approximately, adiabatically enclosed. The first reaction to be considered is an endothermic process, the vaporization of water at its normal boiling temperature (100 "C (373.15 K) and 1 atm pressure): H20(1) = H,O(g)
(8.4)
With the background provided by the previous Section, you should be able to work through this example for yourself, so try the following Question.
A heating coil with a power-rating of 50 W is immersed in water in a Dewar flask, as shown in Figure 8.3: the apparatus is maintained at a constant pressure of 1 atm. When the water boils (at 373.15 K), the amount that vaporizes is measured by condensing it in a weighed flask. With the water boiling in the Dewar flask, 1.62 g of water are condensed over a period of 1 minute and 21 seconds. (a) What is the enthalpy change associated with the vaporization of 1.62 g of water at its normal boiling temperature? (b) What is the enthalpy change for the vaporization of 1 mol of water under these conditions?
Figure 8.3 Apparatus for measuring an enthalpy of vaporization.
The vaporization or melting of pure substances are examples of phase transitions. In principle, the procedure outlined in Question 8.2 can be used to determine AH for such transitions in the endothermic direction. In practice, of course, accurate determinations would use equipment, known generally as a calorimeter. This is much more sophisticated than a simple Dewar flask, but the essential features of the determination would be the same. 56
Figure 8.4
Winter scene showing melting ice, and steam issuing from a subterranean hot spring in which boiling temperatures are sometimes reached. The two phase transitions take place at different but constan t temperatures.
Why is the measurement of the enthalpies of phase transitions relatively straightforward? Phase transitions take place at constant temperature (Figure 8.4), so the electrical work done on the system is a direct measure of AH at that temperature (provided that q = 0, of course!). Using a thermally insulated calorimeter to measure AH for an exothermic reaction is not quite as straightforward, as you will now see. The example we have chosen is the neutralization reaction between an acid and a base: H+(aq) + OH-(aq) = H20(1) (8.5) In a typical experiment, 0.1 litre of HCl(aq) and 0.1 litre of NaOH(aq), both of concentration 2.0 mol litre-l and both at 300 K, were mixed in a thermally insulated calorimeter, at a constant pressure of 1 atm.
If the reaction is exothermic, what do you predict will happen? Since the energy released by the reaction cannot be transferred to the surroundings, the temperature should rise. It does: to 3 11 K in the experiment mentioned above. This change is illustrated schematically in Figure 8.5, where we have used the generalized notation of ‘reactants’ for H+(aq) + OH-(aq) and ‘product’ for H20(1). 57
Figure 8.5 The exothermic reaction H+(aq) + OH-(aq) = H20(1), carried out in a thermally insulated calorimeter.
But how do we relate this temperature change to the required value of AH at a fixed temperature of 300 K? To examine this question, let us call reactants at the initial temperature (300 K) ‘state l’, and product at the final temperature (311 K) ‘state 2’. Then the enthalpy change for the process in Figure 8.5 can be represented as follows : M1+2
= H2 - HI = H(product, 3 11 K) - H(reactants, 300 K)
(8.6)
To determine the value of requires our statement of the first law for a change at constant pressure, Equation 8.1: AH = q
+ We]
(8.1)
Now, in the experiment described earlier, no electrical work is done on the system, so wel = 0. Moreover, if we assume that the calorimeter is perfectly insulated (admittedly an idealization), then the heat transfer will also be zero. Combining these results, wel = 0 and q = 0
A surprising and somewhat alarming conclusion at first sight! It becomes less so once we recognize that the change illustrated in Figure 8.5 is not the one we set out to measure. Indeed, for any experiment like this, will always be, at least approximately, zero. What change are we interested in? Reactants at the initial temperature going to product at the same temperature. If we call product at the initial temperature (300 K) ‘state 3’, then the enthalpy change we are after is AH1+3 = H(product, 300K) - H(reactants, 300K) (8.7) The two enthalpy changes that we have described are illustrated schematically in Figure 8.6. But how can we relate the result of our first experiment to the desired enthalpy change, Remembering that H is a state function, use Figure 8.6 to derive an expression and the enthalpy change between states 2 and 3. in terms of for As H i s a state function, then, with reference to Figure 8.6, we can get from state 1 to state 2 either directly (as in the experiment described earlier) or by way of state 3. In other words, you can complete the cycle in Figure 8.6 by drawing in an arrowfrom state 3 to state 2 (see Figure 8.7): in symbols, AH1+2
58
=
m1-+3
AH3+2
Figure 8.6 Schematic representation of the enthalpy changes AH1 +,2 and m1+,3*
so B 1 + 3
= m 1 + 2 - M3-+2
where B 3 + 2
= H(product, 3 11 K) - H(product, 300 K)
(8.9)
Now concentrate on Equations 8.8 and 8.9. You have just seen that M 1 + 2 is effectively zero: but how can M 3 + 2 be determined? The answer is that a second experiment is required: it’s usually called a calibration experiment. For example, to complete our determination, the product (water, or, more accurately, an aqueous solution of sodium chloride, in this case) is first allowed to cool back to the original temperature, 300 K. When a heating coil with a power-rating of 50 W is immersed in the liquid, and the current is switched on for 220 s, the temperature rises back to 311 K. What is the value of
(Equation 8.9) in this case?
Figure 8.7 The cycle of Figure 8.6 completed by the addition of an arrow from state 3 to state 2. 59
Assuming once again that the system is perfectly insulated, q = 0, so AH3+2
= w,1 = (50 J S-') x 220 s = 11OOOJ = 11.0kJ
What is the measured value of AH for the neutralization reaction in Equation 8.5 at a temperature of 300 K and a pressure of 1 atm? From Equation 8.8, the required enthalpy change is given by: Ml+3
= M 1 + 2 - M3+2
=O-llkJ =-llkJ Notice that AH for this process is, as expected, a negative quantity: the reaction is exothermic. To summarize, the essential elements of this sort of determination are: 1 To perform the reaction under conditions in which the heat flow to or from the system is kept as close to zero as possible: the temperature change resulting from the reaction is measured. 2 To relate this temperature change to the desired enthalpy change: this is achieved by means of a second, calibration, experiment in which a measured amount of electrical energy is transferred to the reaction product(s). This completes our brief discussion of the experimental measurement of enthalpy changes. In practice, many different sorts of calorimeter have been developed for studying different types of reaction (Figure 8.8). Nevertheless, the principles underlying their design and use can always be traced back to the first law of thermodynamics, as for the examples considered here.
As measured in a calorimeter, the value of AH for a reaction depends on the amounts of reactants used in the experiment. To have to state the amounts actually chosen, in addition to the temperature and pressure, would hopelessly complicate the comparison of AH values from different sources. Thus, experimental results are usually quoted as molar enthalpy changes, and denoted by attaching a subscript to the symbol AH, as AH,.
You met one example in the answer and comment to Question 8.2, namely the molar enthalpy of vaporization of water. As you saw, this represents the molar enthalpy change for the process in Equation 8.4: H20m = H20(g) (8.4) where AHm = 44.3 kJ mol-'.
60
Figure 8.8 A simple research calorimeter, showing the Dewar flask used as the reaction vessel. The plastic disc covers the Dewar during measurements, and contains apertures for the stirring facilities (note the pulley wheel) and electrical leads to an internal heating coil and thermistor (a temperature-measuring device). At the top of the frame is the motor that drives the stirrer.
But how about a more complicated reaction, such as the one between sodium and chlorine that you met in Section 5? Na(s)
+
;Cl,(g) = NaCl(s)
(5.2)
In this case, the molar enthalpy change is defined as the enthalpy change when one mole of sodium reacts with half a mole of C12(g)to give one mole of sodium chloride. Now suppose that, in a particular experiment, 2.3 g (0.1 mol) of sodium reacts completely with chlorine, and that the corresponding enthalpy change is found to be AH = -41.12kJ. What is the molar enthalpy change for the reaction in Equation 5.2? When 0.1 mol of sodium reacts completely with chlorine, AH = -41.12 kJ. But we require AH for the case when 1 mol of sodium reacts completely with chlorine. Thus,
If you look back at Question 6.1, you will see that this corresponds to the value of AH quoted for the reaction in Equation 5.2. Indeed, all of the enthalpy changes cited in Sections 6 and 7 are actually values of AHm, and should strictly have the unit kJ mol-I. But there is a more important point about the value of AHrn: it depends on which balanced equation you choose to represent the reaction. The molar enthalpy change of an equation, such as Equation 5.2, is the enthalpy change of the reaction involving a certain number of moles of each substance in the equation. For each substance, that number is the one preceding its formula in the equation. Thus, if we multiply Equation 5.2 through by four, then the value of AHmmust also be quadrupled: 1
Na(s) + ?C12(g) = NaCl(s); AHm = -411.2kJmol-1 4Na(s) + 2C12(g) = 4NaCl(s); AHm = -1 644.8 kJ mol-1
(5.2) (8.10)
This is because the molar enthalpy change of Equation 8.10 refers to the enthalpy change of the reaction in which 4 moles of sodium react with 2 moles of chlorine to give 4 moles of sodium chloride. So, when we quote AHmvalues in kJ mol-l, the ‘per mole’ (mol-l) has a special meaning; it is ‘per mole’ of a chosen, balanced equation, not ‘per mole’ of any particular substance. Consequently, it is meaningful to quote a value of AHm,f o r a reaction only if you know which balanced equation you are talking about.
Use the experimental results quoted in Section 8.2 to calculate AHmfor the following reaction at 300 K and 1 atm pressure: H+(aq) + OH-(aq) = H20(1) (8.5)
61
A calorimeter is used to determine the enthalpy change when 0.50mol of lithium reacts completely with oxygen to give 0.25 mol of lithium oxide, Li20. The value obtained is AH = -149.5 kJ. Calculate and write down the values of AHmfor the following reactions: 2~i(s+ ) ;o2(g) = L ~ ~ o ( s ) 4Li(s) + 02(g) = 2Li20(s) 8Li(s) + 202(g) = 4Li20(s)
-
Energy can be transferred to a system either by heating it or by doing electrical work on it. For a change at constant pressure, our statement of the first law of thermodynamics expresses this result in terms of the enthalpy of the system: AH = q + w,1 Only differences in enthalpy can be determined, never absolute values. Enthalpy is a state function: its value depends only on the state of the system, as specified by its temperature, pressure and composition. For an adiabatically enclosed system (q = 0), our statement of the first law provides a means of measuring: (i) the enthalpy increment, AH = H(T2) - H ( T 1 ) ,for a pure substance, or mixture of substances, at constant pressure; (ii) the enthalpy of transition of a pure substance, at constant temperature and pressure; (iii) when combined with a suitable calibration experiment, the enthalpy change for an exothermic or endothermic reaction at constant temperature and pressure. Now try the following Questions.
A sample of 1.308 g of zinc reacted with an excess of dilute hydrochloric acid in a thermally insulated calorimeter. The temperature of the calorimeter and its contents rose from 298.15 K to 299.5 K. The calorimeter and its contents were then cooled back to 298.15 K. A heating coil, with a power rating of 50 W in the calorimeter, was switched on for 6 1.5 s in order to raise the temperature back to 299.5 K. You may assume that the pressure remained constant throughout., Take the mass of one mole of zinc as 65.38 g. Calculate the value of AHmat 298.15 K for the reactions: (i) Zn(s) + 2H+(aq) = Zn2+(aq) + H2(g) (ii)
LZn(s) 2
+ H+(aq) =
LZn2+(aq) + LH,(g) 2
2
In answering this question, give an indication of any assumptions you make.
The molar enthalpy change for the reaction between aluminium and chlorine is -704.2 kJ mol-l. What is deficient about this statement? 62
If a reaction happens quickly, its enthalpy change can be determined in a calorimeter. But many reactions occur only very slowly, or not at all; how then can we determine their enthalpy changes? Consider first the decomposition of sodium chloride: NaCl(s) = Na(s)
+
C12(g)
(5.1)
As noted in Section 5 , equilibrium lies to the left, and the reaction does not occur. How can its enthalpy change be obtained? The answer is to determine the enthalpy change of the reverse reaction, which happens quickly: Na(s)
+
1
?C12(g) = NaCl(s); AHm= -41 1.2 kJ mo1-I
( 5-2)
The enthalpy change of Reaction 5.1 is minus that of Reaction 5.2; that is, AHm(5.1) = +411.2 kJ mol-I Here we have used the fact that enthalpy is a state function. In Equation 5.1, sodium chloride breaks down into sodium and chlorine; in Equation 5.2, the sodium and chlorine recombine to regenerate sodium chloride. Provided that the sodium chloride that we start with is at the same temperature and pressure as the sodium chloride that we finish with, the overall enthalpy change must have been zero. Thus, Mm(5.1)+ AHm(5.2) = 0 AHm(5.1) = -AHm(5.2) = 411.2kJmol-' (9.1) Unfortunately, the procedure that we have just followed cannot always be used, because of a lack of data. For example, in Activity 2.1, you saw that the reaction between copper and hydrogen ions does not happen: Cu(s) + 2H+(aq) = Cu2+(aq)+ H2(g) (2.9) But if one tries to determine the enthalpy change of this reaction by studying the reverse reaction between hydrogen gas and Cu2+(aq),one finds that this reaction does not occur either. Equilibrium must lie either to the left or the right of Equation 2.9, but the movement towards it must be a very slow process. How then can AHm(2.9)be determined?
Consider the reactions of metallic zinc, firstly with aqueous hydrogen ions and secondly with copper ions, Cu2+(aq).You saw both reactions in Activities 2.1 and 2.2, and both were reasonably fast. Suppose then that we determine the enthalpy of each reaction in a calorimeter. The results are: Zn(s) + 2H+(aq) = Zn2+(aq)+ H2(g); AHm= -153.9 kJ mol-l (9.2) Zn(s) + Cu2+(aq) = Zn2+(aq)+ Cu(s); AHm= -218.7 kJ mol-1 (9.3) 63
Now chemical changes, along with their enthalpy changes, can be added and subtracted. This technique was used in the previous Section. Subtract Equation 9.3 from Equation 9.2. What reaction does the resulting equation represent, and what is its enthalpy change? On subtraction, zinc and its ions disappear: 2H+(aq) - Cu2+(aq) = H2(g) - Cu(s); AHm = 64.8 kJ mol-I
(9.4)
Carrying the formulae with minus signs in front of them over to the other side of the equation, we get Cu(s) + 2H+(aq) = Cu2+(aq)+ H2(g); AHm = 64.8 kJmol-’ (9.5) The result is Equation 2.9, along with the enthalpy change that we were looking for (Figure 9.1). The algebraic manipulation of equations and their enthalpy changes is an application of Hess’s law. This states that:
Like the procedure of Section 9, Hess’s law depends on the fact that enthalpy is a state function, a point that will become clearer in Section 9.3.1. Here, you should particularly note that a database consisting of, say, 100 selected reactions and their enthalpy changes is very fertile, because the manipulation procedure of this Section allows us to calculate the enthalpy changes of many more than 100 reactions from it. We now turn to the question of what type of reaction to put into the database.
When, in the last Section, you subtracted Equation 9.3 from Equation 9.2, the zinc metal cancelled out. But this is legitimate only if the zinc in Equation 9.2 is in the same form as the zinc in Equation 9.3. So if our database of enthalpies of reaction is to be useful, a particular reactant or product must always appear in the same state. We ensure that this is so by specifying its pressure, composition and temperature. Of these, pressure is the least important, because the dependence of enthalpies of reaction on pressure is only slight. All reactants and products are specified as being at a standard pressure of 100kPa. This is only slightly less than the definition of atmospheric pressure (1 atm = 101.325kPa). Next we take composition. For distinct substances, such as the zinc metal and hydrogen gas in Equation 9.2, or the liquid water in Equation 8.4, we specify that the substance must be pure. For dissolved substances, such as the ions H+(aq) and Zn2+(aq)in Equation 9.2, some kind of concentration must be specified as well. This is a complicated business for reasons that we shall not go into; all we note here is that it results in AHmvalues that are appropriate for ions in dilute solutions. This combined specification of pressure and composition gives us what is called the ‘standard state’ of a substance. For example, the standard state of zinc metal is the pure metal at a pressure of 100kPa. When reactants and products are in these standard states, the molar enthalpy of reaction is labelled with a superscript e,thus: AH:. 64
Figure 9.1 Copper does noi: react with dilute hydrochloric acid: in particular, Reaction 9.5 does not law us what its entha1p:y change would be if it did.
This quantity, AH:, is called the standard molar enthalpy change *. The values of AH: now depend only on the temperature. Most compilations, including that in the Data Book (on the CD-ROM), list values of AH: at a temperature of 298.15 K (25 "C). All values of AH: quoted so far in this Book, have, in fact, been AH: values at 298.15 K.
Table 9.1 Standard enthalpies of formation of Some pure substances at 298.151K
What kinds of standard enthalpy change make up our database? One enthalpy change is chosen for each chemical species that is included. It is called the standard enthalpy of formation. You will be dealing with two kinds of standard enthalpy of formation: those of pure substances, and those of aqueous ions.
9.3.1 Enthalpies of formation of pure substances The standard enthalpy of formation of a pure substance, AH: at 298.15 K, is defined as the standard molar enthalpy change at 298.15 K of the reaction in which one formula unit of the substance is formed from its elements in their reference states. To clarify this definition, we shall look at the examples in Table 9.1, which have been taken from the Data Book. For liquid water, Table 9.1 tells us that AH7(H20, 1) = -285.8 kJ mol-l. One formula unit is H20(l), which tells us that the balanced formation reaction (Figure 9.2) is:
and its standard enthalpy change is AH:
= -285.8 kJ mol-'.
Notice that the physical state of the pure substance, liquid water, must be specified explicitly. This explains why there is a second entry for the formula H20 in Table 9.1. Write down the equation to which this second entry refers, along with its standard enthalpy change. The second entry is for gaseous water, H,O(g):
Our definition referred to formation from the reference states of the elements. These are nearly always the commonest form in which the element exists at 298.15 K and 100 kPa -effectively room temperature and pressure. Thus, the diatomic gas is the reference state for hydrogen, oxygen and chlorine, etc., and the liquid is the reference state for mercury and bromine. Some elements exist in more than one crystalline form as solids at room temperature and pressure. The commonest form is then usually chosen as the reference state. For example, graphite rather than diamond is the chosen reference state for carbon. The reference states of the elements can easily be identified in data tables because their AH? values must be zero.
* AH:
is pronounced delta-h-m-standard.
Figure 9-2 Condensed water vapour on a solid surface over which a jet of burning hydrogen gas has been played. The hydrogen flame is coloured red by elements in the gas jet. 65
Why is this? Using H2(g) as an example, by definition, AHy(H2, g) is AH: for the reaction in which H2(g) is formed from hydrogen in its reference state. But this reference state is H2(g), so the formation reaction is
H2(g) = H2(g)
(9.8)
This reaction involves no change whatsoever, so AH:, and therefore LWF(H2, g) is zero. To what reaction does the entry AHF(12, g) = 62.4 kJ mol-1 refer? Iodine is a solid at room temperature and pressure so, as the zero entry in Table 9.1 shows, 12(s)is the reference state. AHy(12,g) refers to the reaction (Figure 9.3): 12(s) = 12(g); AH:
= 62.4 kJ mol-*
(9.9)
Figure 9.3 The modest enthalpy change of Reaction 9.9 shows that the forces that bind one iodine molecule to another in solid iodine are not strong. In a sealed gas jar at room temperature, the purple colour of iodine vapour is visible above small beads of solid iodine.
Write down the reactions to which AH7(Al2O3,s) and AHY(CaCO3, s) refer. Use Table 9.1 to state their standard enthalpy changes.
9.3.2 Calculating standard enthalpies of reaction How do we use our database of values to calculate the standard enthalpy change of any chemical reaction? Let’s do it for the decomposition of calcium carbonate (Figure 9.4): (9.10) CaC03(s) = CaO(s) + C02(g) Table 9.1 gives the
AH? value for each substance in the equation.
Write equations, with enthalpy changes, to which these AH? values refer. See the following table.
66
Figure 9.4 A methane-oxygem torch is used to decompose limestone (calcium carbonate). Quicklime (calcium oxide, CaO) is produced, and at these very high temperatures becomes incandescent, emitting an intense white light. This was the ‘limelight’ used in nineteeth-century theatres.
We now use the technique of Section 9.1: we perform an addition and/or subtraction of these equations and their AH: values, which leaves us with Equation 9.10 and its AH: value. (This is the equivalent of applying Hess’s law.) As the following table shows, this can be done by just reversing Equation 9.11 and then adding the three together:
Although it is always possible to write down a table like this, there is a simpler and more general way. What the table does is to add the formation reactions of CaO(s) and C02(g), and then subtract the formation reaction of CaC03(s).It also does the same thing for the AH? values. Overall then, it adds the AH? values for the products of Equation 9.10, and substracts the AH? value for the reactant:
AH:
= mY(CaO, s)
+
A H ~ ( C O g) ~ , - M?(CaC03, s)
This is a completely general result: it can be used to calculate AH: for any chemical reaction. Thus, for the general reaction, a A + bB + ... = x X + y Y + ...
(9.15)
(9.16)
(where the lower case a, b, x,y, ... etc, are the numbers that precede the formulae also known as ‘coefficients’ -in the balanced chemical equation), we can write: AH: = {xAH:(X) + y M Y ( Y ) + ...} - {aAH?(A) + bAH:(B) + ,..} (9.17) or, making use of the symbol C (capital Greek sigma) to represent the act of adding things together,
A further example should make this clearer.
Use information from Table 9.1 to calculate AH: at 298.15 K for the reaction (9.19) N204(g) = 2N02(g> From Equation 9.18, and noting that the value for the product, NO2, must be multiplied by two: = 2 MY(NO2, g) - M:(N204,
g)
(9.20)
= { 2 x 33.2 - (9.2)) = 57.2 kJ mol-*
67
If you would like some more practice using Equation 9.18 at this stage, then try the following Question.
Use information from the Data Book to calculate AH: at 298.15 K for the following reactions: (i) (ii) (iii) (iv)
1 + zC12(g) = NaCl(s) PbO(s) + C(graphite) = Pb(s) + CO(g)
Na(s)
SOCl2(1)+ H20(1) = S02(g) + 2HCl(g) 3K(s) + AlC13(s) = 3KCl(s) + Al(s)
(5.2) (3.1) (6.4) (3.9)
Unlike the examples in Question 9.2, the reactions that you observed in Activities 2.1 and 2.2 involved aqueous ions as well as neutral species. To extend the procedure represented by Equation 9.18 to reactions of this type requires a definition of AH: for an aqueous ion. However, because ions are charged, this presents problems. Consider the chloride ion, Cl-(aq). Our definition of AH? for pure substances suggests that one possible formation reaction is the formation of the ion from chlorine gas (the reference state of the element) with an extra electron to create the negative charge: (9.21) 1 Cl,(g) + e- = CI-(aq)* Unfortunately, reactions of this kind involve only the formation of a single type of aqueous ion. They cannot be performed in a laboratory without forming ions of opposite charge at the same time, so AHz(9.21) is not experimentally measurable. The best we can do is to study reactions in which two kinds of ion are formed from their constituent elements; for example 2H2(g) 1 + ;c12(g) = H+(aq) + Cl-(aq) (9.22) We can, for example, determine AH: for this reaction by combining hydrogen and chlorine gases (Figure 9.5) in the presence of water. The gases form hydrogen chloride, HCl(g); this then dissolves in the water to give hydrochloric acid, which contains H+(aq) and C1-(aq). The value of AH:(9.22) is found to be -167.2 kJ mol-'. What relationship can reasonably be assumed to exist between M:(9.22), AH: (H+, aq) and AH? (Cl-, aq)?
and
Since, in Equation 9.22, H+(aq) and Cl-(aq) are formed together from the reference states of their elements, it seems reasonable to write M:(9.22)
= AH?(H+, aq)
+
AHy(C1-, aq)
(9.23)
* We again use (aq) to indicate that the ion is in aqueous solution, but you may be puzzled by the absence of any reference to water on the left-hand side of Equation 9.21. Although one could write 'excess water' or '(aq)' on the left-hand side of the equations, it is conventional not to do so. The same is true for equations that you are more familiar with, like Equation 6.3. 68
Figure 9.5 A glass jet of hydrogen burning in a jar of chlorine to give hydrogen chloride, HCl. In the presence of water, the HC1 gals dissolves to give aqueous hydrogeri and chloride ions.
The problem now becomes one of assigning one part of M z ( 9 . 2 2 ) to M ? ( H + , aq), and the rest to AH?(Cl-, aq). To get around this problem, an arbitrary convention is introduced: we simply decide that
Given this convention, and the fact that AHE(9.22) = -167.2 kJ mol-', what is AH~(c~-,aq)? Substitution of Equation 9.24 into Equation 9.23 gives the value AHF(Cl-, aq) = -1 67.2 kTmol-'. By using this convention and the all-important Equation 9.18, we can build up a collection of AH? values for individual aqueous ions. For example, we can now determine AHF(Na+, aq) from the experimental value of the standard molar enthalpy of solution of sodium choride, which was given in Section 6: NaCl(s) = Na+(aq) + Cl-(aq); AH: = 3.9 kJ mol-' (6.3) We have just set M:(Cl-, aq) = -167.2 kJ mol-', and we know that My(NaC1, s) = -411.2kJmol-l. What is M y ( N a + , aq)? Applying Equation 9.18 to Equation 6.3, ME(6.3) = M?(Na+, aq) so
+
AH?(Cl-, aq)
- M?(NaCl, s)
(9.25)
M F ( N a + , aq) = M:(6.3) - M?(Cl-, aq) + AHY(NaC1, s) = {+3.9 - (-167.2) + (-411.2)} kJmol-' = -240.1 kJ mol-'
If you check your Data Book, you will find that this is indeed the value of AHy(Na+, aq). Moreover, by exploiting Equation 9.18, you can now use the database of standard enthapies of formation to calculate the standard enthalpy changes of a host of chemical reactions involving both pure substances and aqueous ions. Finally, it is not difficult to show that the assignment made in Equation 9.24 is equivalent to defining AH? for a possible aqueous ion, such as Na+(aq), as AH: for a reaction of the type: (9.26) Na(s) + H+(aq) = Na+(aq) + +H,(g) Similar formation reactions can be written for other aqueous ions, but we defer further discussion of this until Section 11.2.
69
1
2
3
4
Standard enthalpies of reaction are molar enthalpy changes of reactions that take place at constant temperature, and in which reactants and products are in their standard states. For distinct substances, these standard states are the pure substance at a pressure of 100 kPa (0.987 atm). The constant temperature is usually defined as 298.15 K. The standard enthalpy of formation of a pure substance is the standard molar enthalpy change of the reaction in which one formula unit of the substance is formed from its elements in their reference states at a constant temperature (usually 298.15 K). The standard molar enthalpy changes of reactions can be calculated from a database of standard enthalpies of formation by applying Hess’s law using the equation AH: = C AHy(products) - C AHy(reactants) (9.18) By using the convention M Y ( H + , aq) = 0, standard enthalpies of formation of aqueous ions can be included in the database. Equation 9.18 then provides the standard enthalpies of reactions that include ions.
Using information from the Data Book, calculate AH: at 298.15 K for the following reactions: (i) (ii) (iii)
+ 2H+(aq) = Zn2+(aq) + H2(g) Mg(s) + 2Ag+(aq) = Mg2+(aq)+ 2Ag(s) H+(aq) + OH-(aq) = H20(1)
Zn(s)
Like sodium, magnesium burns in chlorine: Mg(s) + C12(g) = MgCl,(s); AH: (298.15 K) = -641.3 kJ m o t 1
(2.4) (2.13) (8-5)
(9.27)
The standard molar enthalpy of solution of MgCl,(s) is -160.0 kJ mol-I, and AH:(Cl-, aq) is -167.2 kJ mol-l. What is the standard enthalpy of formation of Mg2+(aq)?
70
At this stage, a reminder of where we stand may be useful. In Section 7, we used the second law of thermodynamics to find a quantity, AH - TAS, which does for chemical reactions at constant temperature and pressure, what potential energy does for the motion of a stone in a gravitational field. The quantity consists of two terms. We then set out to show how each of them could be experimentally determined. In Sections 8 and 9, you were shown how the first term, AH,could be measured, and how the measurements could be used to build up a database from which other AH values could be calculated. It now remains to repeat the process for the second term, -TAS. The crucial task here is the determination of the entropy change in a reaction, AS, and it is to this that we now turn.
To find out how to determine entropies of reaction, we return to the definitions given in Section 7.1. For this purpose, cases 1 and 3 are the important ones: 1 When heat is transferred to or from a pure substance at constant temperature and pressure, the change in the entropy of the substance is
3
When a chemical reaction takes place at constant temperature and pressure, the change in entropy associated with the reaction is given by ASreac.= S(products) - S(reactants) (7 -4)
Before going further, notice that the definition in Equation 7.4 implies that entropy, like enthalpy (compare Equation 6.2, p. 42) is a state function. In other words, the change in entropy of a system depends only on its initial and final states, and not at all on how it proceeded from one to the other. Now, as we said in Section 7.1, the use of Equation 7.4 depends on the determination of the entropies of pure substances via the expression in Equation 7.2. So we shall start with the entropies of pure substances.
You may be struck by one disturbing feature of the definition centred on Equation 7.2. It implies the transfer of heat to a substance without changing its temperature. In fact this is possible, but only under special circumstances. In what type of process, studied earlier in the Book, was this condition satisfied?
71
When heat is transferred to a substance undergoing an endothermic phase change, such as melting or boiling, the change takes place at constant temperature. One example (Section 8.2) is the vaporization of water at its boiling temperature, 100 "C (373.15 K at 1 atm pressure):
In this case, Tis constant at 373.15 K, so we can calculate AS from Equation 7.2 if we can determine q, the heat that must be transferred to a sample of boiling water to convert it into steam. Our statement of the first law of thermodynamics enables us to do this:
AH = q + w,1
(8.1)
If the conversion from water to steam were brought about by heat transfer (Figure 10.1, top), no electrical work would be done, so AH = q. Alternatively, we could insulate the boiling water in, say, a Dewar flask and carry out the conversion by doing electrical work with an electrical heating element immersed in the water (Figure 10.1, bottom). This time, q = 0, and AH = wel, a relation that was used to determine AH in Question 8.2. Since enthalpy is a function of state, and the same change occurs in both cases, the two AH values are the same. This means that q, the heat supplied in the first conversion, is equal to wel, the electrical work done in the second. In this case therefore: 4 = We1
(10.1)
and, from Equation 7.2, (10.2) T This equation holds only under the very special circumstances outlined above. In particular, the temperature T must not change, and no heat must be transferred into or out of the calorimeter. Under these conditions, all the electrical work done goes to bring about the desired phase change.
AS =
We1
-
Figure 10.1 Two ways of convwting 1 mole of water at the boiling point into steam. At the top, no electrical work is done so AH = q; at the bottom, the sample is evaporated by electrical work in an insulated container when q = 0, so AH = w,l. As the two values of AH are identical, q = wel. 72
But the example in Question 8.2 highlights a further point. Evidently, one experiment suffices to determine both the enthalpy and entropy of vaporization for water (or any other pure substance), since, in the second of our two conversions (Figure 10.1, bottom), AHva, = we1.Thus, AH,,, and AS,,, are related by a special case of Equation 10.2: ( 10.3)
* vap where Tvap is the normal boiling temperature. From the results in the answer to Question 8.2, the molar entropy of vaporization of water is then: 44.3 kJ mol-' = 373.15K
-
44.3 x lo3 Jmol-' 373.15 K
= 118.7 J K-' mol-1
Equation 10.3 is one example of a general relation between entropies and enthalpies of transition for pure substances. Now try the following Question. This is important!
Solid chlorine can be made by cooling chlorine gas to, say, 150 K. On warming, the solid fuses (melts) at 172 K, when AHfu,= 6.4 kJ mol-l. Further warming causes the liquid to boil (vaporize) at 239 K when AHvap= 20.4 kJ mol-l. One mole of chlorine (C12) has a mass of 70.9 g. Calculate: (a) the molar entropy of fusion of solid chlorine; (b) the molar entropy of vaporization of liquid chlorine; (c) the entropy of vaporization of 7.09 g of liquid chlorine.
In this Section, we shall seek a way of determining the entropy change of a substance when its temperature is increased. We shall use chlorine as an example, finding the entropy change when a sample of this substance is warmed from the absolute zero, OK, to 298.15 K (25OC) Two important events happen during this temperature change: what are they? At 0 K, the chlorine is solid; it fuses at 172 K, and then vaporizes at 239 K (Question 10.1). It follows that the entropy change of chlorine when it is warmed from 0 K to 298.15 K is the sum of five parts: 1 The entropy change of the solid chlorine when it is warmed from 0 K to the fusion temperature of 172 K. 2 The entropy of fusion of the solid chlorine at 172 K. 3 The entropy change of the liquid chlorine when it is warmed from 172 K to the vaporization temperature of 239 K. 4 The entropy of vaporization of the liquid chlorine at 239 K. 5 The entropy change of the gaseous chlorine when it is warmed from 239 K to 298.15 K. 73
In Question 10.1, you calculated parts 2 and 4; it remains to calculate parts 1, 3 and 5. In all three cases, the central problem is that of finding the entropy change when the temperature of the chlorine rises from a lower value, T I ,to a higher value, T2. The first step is to determine q, the heat needed to bring about the temperature increase. The experimental procedure is similar to the one that we used when determining the entropy of a phase transition, and is shown in Figure 10.2. We insulate the chlorine in a calorimeter, and determine the electrical work, wel, that is needed to bring about the required temperature change. The heat that will induce the same temperature increase when no electrical work is done can then be found from the equation 9 = We1 (10.1)
Figure 10.2
How to determine the heat q required to raise the temperature of one mole of solid chlorine from T I to T2 at constant pressure. On the lower path, the job is done Iby a measured amount of electrical work in an insulated container in which q = 0, so AH = wel. On the top path, the job is done by heat transfer, so weI = 0 and AH = 4. Thus, = wel. Let us now consider an experiment in which a small temperature increase of 1.10 K is induced in a sample of solid chlorine at the very low temperature of 13.50 K. The results are shown in Table 10.1, and the value of wel tells us that q = 3.728 J. Table 10.1 Data for an experiment in which electrical work raises the temperature of one mole of solid chlorine from T I (= 13.50 K) to T2 (= 14.60K)
But now we encounter the difficulty mentioned at the beginning of Section 10.1: Equation 7.2 shows that the entropy change is defined through a heat transfer at a constant temperature T:
74
Yet the temperature has varied between 13.50 K and 14.60 K, so how can the equation be used? If a single temperature must go into Equation 7.2, then the most sensible choice would seem to be the average of T I and T2, which has been denoted T’. The small overall increase encourages this measure, because throughout the change, the temperature never differs by more than k0.55 K from the mean value of 14.05 K. We can now use Equation 7.2 to calculate an approximate entropy change, AS’, by using the equation
A S‘ = -4
= w,1 ( 10.4) T’ T’ As Table 10.1 records, this gives a value of 0.265 J K-’ mol-l. Now, this suggests a way of estimating AS for a much bigger temperature change. Having raised the temperature to 14.60 K, we now induce a whole succession of further small temperature increases -from 14.60 K to 15.82 K, then from 15.82 K to 18.98 K, and so on. In Table 10.2, such a succession of increments takes the temperature from the original 13.50 K up to 90.00 K.
In all but one case, the approximate entropy change, AS’, has been calculated from Equation 10.4. We could then add up all the AS’ values to get an estimate of CAS’, the total entropy change between 13.50 K and 90.00 K at 1 atm pressure: (10.5) S(C12, S, 90.00K) - S(C12, S, 13.50K) CAS’
Now try the following Question. (a) To confirm that you have followed the argument, use the data in Table 10.2 to calculate the missing AS’ value for the increment where T I is 20.64 K and T2 is 25.22 K. (b) Now use Equation 10.5 to obtain a value of the entropy difference between one mole of chlorine at 90.00 K and at 13.50 K. For the solid chlorine of Question 10.2, the summation process yields a total entropy change of 39.39 J K-’ mol-l. As Equation 10.5 implies, and as we have emphasized, this value is approximate, but there are mathematical techniques for improving it. To see how, look at the AS’ values for the increments in column 6 of Table 10.2. They display no clear trend, sometimes rising, and sometimes falling as one moves down the column. Why does this happen? It is because the changes in temperature, AT, vary irregularly. When AT is large, AS’ is large; when AT is small, AS’ is small. This suggests that we will observe a more regular pattern if we concentrate on AS’/AT. In the last column of Table 10.2, the values of AS’/AT have been calculated for all but one of the increments. Fill in the missing value, using your answer to Question 10.2a. What pattern do you observe in the AS’/AT values? The missing figure is 0.450 J K-2. Values of AS’/AT now increase steadily, reaching a maximum in the region of T’ = 40 K; thereafter they decline.
7s
Table 10.2 Data for an experiment in which successive amounts of electrical work raise the temperature of one mole of solid chlorine from 13.50K to 90.00 K in a succession of small increments*
* Note that a set of actual experimental results would show the electrical work done on the system as a whole: generally this would comprise a calorimeter and all its contents, including a thermometer and heating coil as well as the chlorine. The electrical work required to increase the temperature of the calorimeter, thermometer and heating coil is determined in a separate calibration experiment with an empty calorimeter. We have made allowance for everything except the chlorine in Table 10.2 by subtracting the electrical work done in this calibration experiment.
This trend is more clearly revealed by the plot of AS’/AT against T’ in Figure 10.3. Here, the important point is that because AS’/AT shows a regular trend with temperature, a curve can be confidently drawn through the data points. Although the AS’ values were obtained by an approximation, the curve-drawing process diminishes the effects of the errors in individual points. Moreover, the curve provides AS’/AT at any value of 7‘ between 13.50 K and 90.00 K, and this means that small temperature increments other than those in Table 10.2 can be studied. We now take advantage of this by imagining an increment covering a very narrow temperature range. This increment has been drawn as a narrow vertical strip in the inset of Figure 10.4. The left-hand and right-hand verticals mark the initial and final temperatures, T I and T2,respectively, and the width of the strip is AT. These features appear most clearly in the enlargements of the places where the strip intersects the 76
Figure 10.3 A plot of the values of AS’/AT against T’ usmg the data on solid chlorine in Table 10.2.
curve and the T’ axis. The mean temperature T’ is marked by the central vertical, and intersects the curve at the value of AS’/AT for the increment. To obtain the value of AS’ for the increment, therefore, one simply multiplies this value of AS’/AT by AT.
Figure 10.4 A tiny temperature increment for one mole of solid chlorine can be repre-sented by drawing a very narrow vertical strip in Figure 10.3, the vertical sides being at the lower and upper temperatures, T , and T2 Here the intersections with the curve and the temperature axis have been magnified.
So what area in Figure 10.4 is AS’ equal to? The rectangle DPRE has an area DP x PR. As DP = BQ = AS’/AT and AT = PR, this area is equal to AS’. Now this would be true whatever the width of the increment, but now for something that is not. If the increment is very narrow, the tiny section of curve AC that crosses it can be thought of as a straight line. Consequently, as B is the mid-point of both AC and DE, the area of the triangle BCE is equal to that of the triangle BAD. So what area besides the rectangle DPRE is also equal to AS’?
That of the strip ACRP beneath the curve. It can be obtained by adding BCE to the rectangle DPRE, and subtracting BAD. As the areas of the two triangles are the same, the rectangle and the strip have the same area.
Now suppose the sample of solid chlorine is taken through a much bigger increment -for example between the temperatures 20 K and 80 K in Table 10.2. These boundary temperatures can be represented by vertical lines from the horizontal axis to the curve in Figure 10.4. We can now imagine the temperature being increased from 20 K to 80 K by a whole succession of many tiny increments. The argument that we have just given shows that the entropy change for each tiny increment is the area of the strip beneath the curve bounded by the lower and upper temperatures. The entropy change for the overall large increment is the sum of the areas of these strips, and this is simply the area beneath the curve between the vertical lines at 20 K and 80 K. In Figure 10.5, the curve of Figure 10.3 has been replotted with a dotted extension from 14.05 K, the lowest value of T’ in Table 10.2, back to the absolute zero. Such an extension is always necessary in entropy measurement because the absolute zero is unattainable. It has been calculated by using a theoretical equation whose basis we do not discuss, but as you can see, it implies that AS’/AT is zero at 0 K.
Figure 10.5 The plot of AS’/AT against T for one mole of solid chlorine over the range 0-90 K. The extension of the experimental data from 14 K back to 0 K has been made using a theoretical equation. The area beneath the curve has been subdivided into approximate triangles and rectangles. 78
To check that you are following the argument, try the following Question.
Use Figure 10.5 to determine the molar entropy change for solid chlorine when its temperature rises from 0 K to 90 K. Assume that regions a and b in Figure 10.5 are rectangles, and that regions c, d and e are triangles whose area is base x height 2
To summarize, the entropy change for a pure substance over a temperature range that does not include a phase change can be determined from values of the quantity
AT
we1 T ‘AT
(10.6)
which is obtained for a succession of small increments of about 1-5 K, T’ being the mean temperature of the increments, and wel the electrical work needed to induce the temperature rise, AT. These values are plotted against T’; the entropy change is then the area beneath the curve between the lower and upper limits of the temperature range. In Figure 10.5, we have dropped the superscript dashes against AS and Ton the axes because we assume that the curve we derive will give us accurate values of ASIAT for very small temperature increments centred on a temperature T. Research workers who determine entropies perform the calculation mathematically, computing the area by a technique known as integration. Such workers write the entropy change when a solid substance increases its temperature from T I to T2,in the form: (10.7) where C, is the heat capacity of the substance, and the symbol ) marks an integration process. Nevertheless, both this method, and the one we have used, amount to the same thing.
By using measurements of the type listed in Table 10.2, and the technique used in answering Question 10.3, the entropy increment for any pure substance going from one temperature to another can be determined. But we can go a stage further. Unlike enthalpies, it is possible to determine an absolute value for the entropy of a pure substance at a given temperature. This absolute scale is based on a third, and final, ‘rule of the game’ -a proposition known as the third law of thermodynamics. This law is adopted because it establishes useful relationships between important ideas in thermodynamics and in quantum mechanics. For our purposes, we shall interpret the law as saying:
* The absolute zero of temperature cannot actually be obtained, but it can be approached (experimentally): this is why we write T + 0. 79
If a pure solid substance undergoes no phase change between 0 and 298.15 K, its absolute entropy at 298.15 K can now be determined from a plot like that in Figure 10.5. We know that the area under the graph of AS’IAT against T between 0 and 298.15 K is given by area = S(298.15 K) - S(T+ 0) But the third law tells us that S(T
-+
(10.8)
0) is zero, so Equation 10.8 becomes
area = S(298.15 K)
(1 0.9)
In other words, the area under the graph represents the absolute entropy of the substance at 298.15 K. You should be able to see that the value you obtained in answering Question 10.3 represents the absolute molar entropy of solid chlorine at 90 K. As with enthalpies, calorimetrically determined entropies must be adjusted to a standard reference state, and this requires the specification of a standard value, p*, of pressure: once again the value chosen is 100kPa (or one bar), which is a typical atmospheric pressure. Strictly speaking, the symbol for a standard absolute molar entropy should be written S g together with a specification of the temperature. However, the subscript m is generally omitted where doing so does not cause any confusion. Indeed, it is normal to refer simply to ‘absolute entropies’.
At the beginning of Section 10.2 we set ourselves the task of determining the entropy change AS of a sample of chlorine when it is warmed from 0 K to 298.15 K. The third law of thermodynamics now tells us that this value of AS is identical with of chlorine at 298.15 K. If chlorine underwent no phase the absolute entropy, p, change between 0 K and 298.15 K, the absolute entropy could be obtained by measuring the area under the graph of AS’/AT against T; in Question 10.3, P ( C l 2 , s) at 90 K was obtained in just this way. However, chlorine melts at 172 K and vaporizes at 239 K. Let us see how this affects the entropy determination by examining the complete plot of AYAT against T for chlorine between 0 and 298.15 K, which is shown in Figure 10.6. What is the significance of the breaks at 172 K and 239 K? The one at 172 K marks the fusion temperature, where solid changes to liquid; the one at 239 K marks the vaporization temperature, where liquid changes to gas. The fusion and vaporization breaks divide the area beneath the curve into three parts labelled 1, 3 and 5. Area 1 represents the entropy change when the mole of solid chlorine is warmed from 0 K to 172 K; area 3, the entropy change when one mole of liquid chlorine is warmed from 172 K to 239 K, and area 5 the entropy change when one mole of gaseous chlorine is warmed from 239 K to 298.15 K. Is the sum of these three areas equal to the value of the absolute entropy of chlorine gas, p(Cl,, g) at 298.15 K?
80
TI K
Figure 10.6 The variation of ASIAT with temperature for one mole of chlorine in the range from T -+ 0 to T = 298.15 K. The points mark the limits of the temperature range over which experimental measurements are made.
No; look back to the beginning of Section 10.3, where we divided the entropy change when one mole of chlorine is warmed from 0 to 298.15 K into five stages; areas 1, 3 and 5 only give us the entropy changes for stages 1, 3 and 5 . The calculation of P ( C l 2 , g) at 298.15 K can therefore be completed by adding the entropy changes for steps 2 and 4. These are the AS values for the fusion and vaporization of the chlorine, which were calculated in Question 10.I . The addition for the five stages is shown in Table 10.3; the final value of P ( C l 2 , g) at 298.15 K is 223.1 J K-I mol-I, as given in the Data Book (on the CD-ROM associated with this Book). Table 10.3 The absolute entropy of one mole of C12 at 298.15 K and 100 kPa is obtained as a sum of the entropy changes of five stages as the chlorine sample is warmed from T + 0 to T = 298.15 K
81
In general then, the absolute entropy of a pure substance at a temperature T may be obtained by measuring the area beneath the graph of AS/AT against T between 0 K and T, and adding the terms AHf,,/Tf,, and AHvadTv,p, where these are relevant. Notice that this extension carries with it an important implication: the absolute entropy of a gas will contain contributions from the entropies of fusion and vaporization. Not surprisingly, therefore, the entropies of gases are larger than those of comparable liquids or solids. We shall return to this point in the next Section. In summary: once the necessary measurements and calculations have been performed, the resulting values of p are listed in the chemical literature. A compilation of such values at 298.15 K is included in the Data Book. A quick glance through these values should confirm the generalization we made above. For instance, the values in Table 10.4 confirm the tendency for the entropies of gases to be larger than those of solids or liquids.
Table 10.4 The absolute entropies of gases at 298.15 K are usually larger than those of solids or liquids, as these values show
Among the entries in the Data Book, you will also find values for individual aqueous ions. Like enthalpy data, these rest on the arbitrary assignment of a zero value to one ion: again, the hydrogen ion is chosen, so (by definition) ( 10.10) F ( H + , aq, 298.15 K) = 0 This is another example of an arbitrary convention *; once again, the values so obtained are internally consistent.
The values of p in the Data Book can be used to calculate the standard molar entropy change for a reaction, AS:, by substituting them into a more precise version of Equation 7.4 (Section 7.1):
Using information from the Data Book, calculate AS: reactions at 298.15 K:
for each of the following
(ii) (iii)
1 + ?C12(g) = NaCl(s) Mg(s) + ;O2(g) = MgO(s) CaC03(s) = CaO(s) + C02(g)
(2.5) (9.10)
(iv)
N204(g)
= 2N02(g>
(9.19)
(a) (i)
(b) (v) (vi)
Na(s)
+ 2H+(aq) = Zn2+(aq) + H2(g) Cu(s) + 2H+(aq) = Cu2+(aq)+ H2(g)
Zn(s)
(5-2)
(2.4) (2*9)
The examples in Question 10.4 underline important differences between entropy and enthalpy data-points that will be taken up in the next Section. But closer inspection of 82
* It is only because of this arbitrary convention that the absolute entropies of some aqueous ions have negative values; for example, P ( M g 2 + , aq) = -138 J K-I mol-*.
the results from Question 10.4a raises a further point. In each case the sign of AS: for the reaction can be correlated with whether it results in an increase or a decrease in the number of moles of gaseous species. This observation, which is supported by the more extensive compilation in Table 10.5, harks back to our earlier comment about the large values for gases. It suggests an important and useful generalization:
Table 10.5 Values of AS:
at 298.15 K for a variety of reactions
* Note that chlorine atoms are involved in this reaction. But note the qualification in this generalization: it is restricted to changes in which every reactant and product is a pure substance. In the reactions in Question 10.4b, 0 hydrogen gas is produced, but AS, is negative! Reactions in solution, especially those involving aqueous ions, are distinctly more complex, and there is then no simple way of predicting the sign of AS: with any certainty.
1
2
3
4
5
The entropy of transition of a pure substance is AH/T, where AH is the enthalpy of transition at the transition temperature T. The entropy change of a pure substance when its temperature is raised from T I to T2 can be determined from values of the quantity wel / T ’AT obtained for a succession of small increments, AT, of about 1-5 K, T’ being the mean temperature of the increment, and wel the electrical work needed to induce the increment AT. These values are plotted against T ’;the required entropy change is the area beneath the curve between T I and T2,plus values of M I T f o r any phase transitions in this range. The procedure of point 2 allows absolute entropies to be determined, because according to the third law of thermodynamics, when the temperature of a substance is raised from 0 K to T, the entropy change is equal to the absolute entropy of the substance at temperature T. The entropy change of a reaction, AS:, can be obtained from the absolute by using the equation entropies, p, A S : = Cp(products) - Cp(reactants) (10.11) Because the entropies of gases are so large, a reaction between pure substances that results in an increase in the number of moles of gaseous species will nearly always have a positive value of AS:. 83
In Section 7, we used the second law of thermodynamics to show that, for a spontaneous change at constant temperature and pressure, AH - TAS < 0
(7.10)
In Sections 8 and 9, you were shown how AH could be determined, and in Section 10 how the same could be done for AS. The values of AH and AS then allow you to predict whether or not a given reaction can occur, without ever trying it in practice. Such is the power of thermodynamics! The two-term quantity, AH - TAS, is so important that it deserves a symbol of its own. We define a quantity G in terms of enthalpy and entropy, writing it G=H-TS (11.1)
G is known variously as the Gibbs function or the Gibbs free energy, after the American physicist, Josiah Willard Gibbs (Figure 11.1). According to this definition, what is the SI unit of G? The SI unit of H is J, and that of TS is K x J K-’ = J, so the SI unit of G is the joule.
Figure 11.1 Josiah Willard Gjbbs (1839-1903). *part from three years study in Thus, a change in the Gibbs function at a constant temperature T (and implicitly at Europe during his twenties, this constant pressure) is given by taciturn bachelor lived his life in AG = AH - TAS -2) a house close to Yale University, where he was first a student and then Comparing this result with the inequality 7.10, leads to the following simplified an unsalaried pro-fessor.In 1863, he criterion for a spontaneous reaction: received the first US engineering (11.3) doctorate for a thesis on-the teeth of AG < 0 (at constant T and p ) wheels in spur gearing. During the Does this inequality accord with the intuition about an ‘uphill-downhill’ 187Os, he applied the second law of thermodynamics to chemical character to chemical reactions, as developed in Section 5? equilibrium in a series of papers Yes. In terms of our simple analogy, it seems that spontaneous reactions really which later revolutionized physical do ‘roll downhill’, but on a landscape where ‘altitude’ is measured by the Gibbs chemistry. Johns liopkins University function, not the enthalpy (as in Thomsen’s hypothesis). at Baltimore then offered him a job, and Yale University at last produced Notice that when reactants and products are in their standard states, AG is written a salary to retain his allegiance.
AG*, and the standard molar Gibbs function change is written AGE. Thus, the values of AH: and AS: calculated from tabulated data are strictly related to AG:, as defined by an expression analogous to Equation 11.2: (11.4) AG: = AH: - T A S ~
84
Now try the following Question.
(a) Use information from the Data Book to calculate the standard molar entropy change of the reaction SOC12(1) + H20(1) = S02(g) + 2HCl(g) (6.4) (b) The value of AH: for this process was given in Section 6 as 47.1 kJ mol-l. Calculate the value of AGE. (c) Are the values of AH,: AS: and AGE in accord with the observation that Reaction 6.4 occurs readily at room temperature and pressure?
To adopt the criterion AG: < 0 as the mark of a thermodynamically favourable reaction is to select a special case of inequality 11.3. The great advantage of concentrating on AGE is that this quantity is directly related to the equilibrium constant, K, the other measure of how favourable a reaction is. When a reaction is very favourable, is the equilibrium constant K large or small? It is large; for example, for a general reaction such as A+B=C+D [CI [Dl [A1P I If the reaction is favourable, the products C and D must predominate at equilibrium, so K must be large. K =
~
(11.5) (11.6)
From the discussion above, AG: must be negative for a reaction like this: conversely, if AG* is positive, then K must be small. These observations suggest strongly that 0m AGm and K must be related in some way. Unfortunately, the derivation of this important relation lies beyond the scope of this Book, so we must ask you to take on trust that it is as follows: = -RTlnK (11.7)* = -2.303RTlog K (11.8)*
AGE
where R is a constant, known as the gas constant, which has the value R = 8.314 J K-'mo1-l. The term In K is the natural logarithm of K; log K is its logarithm to the base 10. Although we cannot prove the expressions in Equations 1 1.7 and 11.8 here, they certainly fulfil the conditions outlined above. Consider, for example, the values of log K and of K for a range of values of AGE given in Table 11.1. In particular, notice that when AG: is negative, K is greater than one and, when AG: is positive, K is less than one.
* See Maths Help Box overleaf.
as
Table 11.1 Some numerical values of AG: with corresponding values of log K and K at 298.15 K (from Equation 11.8, with R = 8.314JK-1 mol-l)
Consider a simple isomerization reaction of the type A = B. If AG: proportion of the reactant, A, remains at equilibrium?
= 0, what
It seems, then, that AG: is really just a measure of the equilibrium constant for a reaction at a particular temperature. Table 11.1 shows that the more negative AG: is, the larger is K and so the more favourable is the reaction. Indeed, the value of AG: at 298.15 K need be only moderately negative (around -50 kJ mol-l) to signify a reaction for which the equilibrium position strongly favours the products. By contrast, an equally positive value of AG: indicates a reaction for which little reactant has been converted by the time equilibrium is attained.
86
To reinforce this idea, try the following Question.
Use Equation 11.4 to calculate AG,8 at 298.15 K for each of the following reactions: (i) Na(s) + ;Cl2(g) = NaCl(s) (5.2)
+ 2H+(aq) = Cu2+(aq)+ H2(g) Zn(s) + 2H+(aq) = Zn2+(aq)+ H2(g)
(ii) Cu(s)
(2.9)
(ii)
(2.4)
Decide whether or not the reactions are thermodynamically favourable under standard conditions at 298.15 K.
We are now almost ready for the promised thermodynamic analysis of the hypothesis raised in Sections 4 - 4.2. However, before we embark on it, you must be able to exploit the thermodynamic database in the Data Book; this Section should ensure that this is the case. The database consists of a table, which lists three thermofor a wide range of chemical substances. dynamic quantities, AH?, AG? and A small selection is given in Table 11.2. Two kinds of material are represented: substances that have no overall charge, and aqueous ions. Let us consider substances with no overall charge first. Sodium chloride, NaCl(s), is a good example. Write the equation to which AH? for NaCl(s) refers. Section 9.3.1 identified AH? for a pure substance as the value of AH: for the reaction in which the substance is formed from its elements in their standard reference states: Na(s)
+
+C12(g) = NaCl(s)
(5.2)
Table 11.2 Examples of thermodynamic data at 298.15 K from the Data Book
87
You can see that AGF also carries a subscript f, and this marks the fact that it too refers to this type of reaction. For an uncharged substance, AG?, the standard Gibbs function of formation, is the value of AGE for the reaction in which the substance is formed from its elements as in Equation 5.2. As with AH? values, this means that AG? for any element in its standard reference state is zero. Values of AG? can now be obtained by using the equation A G =~ AH? - TAS? (11.9) which is just a special case of Equation 11.4. This expression, however, contains a trap for the unwary, so be careful. The entropy To change in Equation 11.9 is AS?, but the tables contain absolute entropies p. use Equation 11.9, you must calculate AS? for the substance from the values of s4 in the data base. Use the
values in Table 1 1.2 to calculate AS? for NaCl(s).
AS? refers to the formation reaction: Na(s) +fC12(g) = NaCl(s)
So using Equation 10.11: AS? = p(NaC1, s) - S@(Na,s) - k p ( C l , , g) 1 x 223.1)JK-'mol-' = (72.1 - 51.2 = -90.7 J K-I mol-I As AH? = -41 1.2 kJ mol-I, Equation 11.9 tells us that: AGy = -411.2 kJ mol-' - (298.15 K) x (-90.7 J K-* mol-') Remember to allow for the fact that joules appear in the units of AS and kilojoules in the units of AH: AG? = -41 1.2 kJ mol-' + 27 040 J mol-1 = -41 1.2 kJ mol-I + 27.0 kJ mol-I = -384.2 kJ mol-l This is the value given in the Tables, and the calculation establishes the nature of the for a pure substance. relationship between AH?, AG? and In the case of aqueous ions, we have not yet defined the formation reaction explicitly. Instead, in Section 9.3.3, we obtained AH? values for individual ions by assigning an arbitrary value to one ion. This assignment was made by defining AH?(H+, aq) = o (9.24) However, it is not difficult to show that this assignment is equivalent to defining AH? for a positive aqueous ion as AH: for the reaction in which the ion is formed by the reaction of its elements in their standard reference states with H+(aq), the other product being hydrogen gas. Thus, for zinc, this reaction is (2.4) Zn(s) + 2H+(aq) = Zn2+(aq) + H2(g) Equation 9.18 then tells us that AH: = AH?(z~~+,aq) + AH?(H*, g) - m y ( Z n , S) - ~ A H ~ ( Haq) +,
88
But by definition, values of AH?, for H2(g), Zn(s) and H+(aq) are zero, so
AH$ = A H ? ( Z ~ ~ aq) +, For negative aqueous ions, the formation reaction is similarly defined, but now H+(aq) goes on the right and hydrogen gas on the left; for example 1
1
,C12(g) + ,H2(g) = W a q ) + H+(aq)
(9.22)
Again, all values of AH? are zero by definition, except the one for C1-(aq). Write the formation reaction for S2-(aq) The equation is
S(S) + H2(g) = S2-(aq) + 2H+(aq)
(11.10)
Like values of my,values of p (as you saw in Section 10.5) and of AG? require arbitrary assignments to one ion before a set of individual values can be recorded. Again, zero values are assigned to the hydrogen ion. Hence we arbitrarily say that: F ( H + , aq, 298.15 K) = 0 AG?(H+, aq, 298.15 K) = 0
(10.10) (1 1.11)
This ensures that, for aqueous ions as for other substances, AG? and AH? refer to the same formation reaction, and that they are, respectively, the values of AG: and AH: for that formation reaction. As for pure substances, when performing calculations with Equation 11.9, you must remember that the equation includes AS?, and not p.
Use the
values in Table 11.2 to calculate AS? for Na+(aq).
The formation reaction is: Na(s) + H+(aq) = Na+(aq) + kH2(g) AS? = p ( N a + , aq) = (59.0
+
+
(9.26)
i p ( H 2 , g) - p ( N a , s) - p ( H + , aq)
x 130.7 - 5 1.2 - 0) J K-I mol-1
= 73.2 J K-' mol-1
As AH? = -240.1 kJ mol-', Equation 11.9 tells us that: AG? = -240.1 kJ mol-1 - (298.15 K) x 73.2 J K-* mol-* = -261.9 kJ mo1-l
Again, this is the value given in the Tables.
89
In Question 11.1, you calculated AG: for the reaction SOCl2(1) + H,O(l) = SO,(g) + 2HCl(g)
(6-4)
The value was obtained by computing the values of AH: and AS: using Equations 9.18 and 10.11, and then substituting them into Equation 11.4. Now repeat the calculation using values of AG? from the Data Book and Equation 11.12. Check that you get the same answer.
1
2
3
4
5
6
The standard Gibbs function change for a chemical reaction at a constant temperature Tis given by (11.4) AG: = AH: - T A S ~ At constant temperature and pressure, the criterion for a spontaneous change is AG < 0. We take the criterion AG: < 0 as a definition of a thermodynamically favourable reaction; conversely, a reaction is considered to be unfavourable if A G >~ 0. At constant temperature, the more negative is AG:, the larger is the equilibrium constant, and the more favourable is the reaction. The Data Book (on the CD-ROM) contains a table of AH? , AG? and values for chemical substances. For pure substances, AH? and AG? refer to a formation reaction in which the substance is formed from its elements in their standard reference states. For aqueous ions, the formation reaction takes one of two forms: formation of cations from their elements and H+(aq)with production of hydrogen, or formation of anions from their elements and H2(g)with production of H+(aq). AH? and AG? are related by a special case of Equation 11.4: AG? = - TAS? (11.9) To use this equation, AS: values have to be calculated from the values in tables. AG: values for balanced chemical equations can be obtained from (1 1.12) AG: = C AG? (products) - C AG? (reactants)
This question is concerned with the information in Table 11.3. Taking any further information you require from the Data Book, fill in the two blank entries in Table 11.3. Is the formation of Sc203(s)from the elements thermodynamically favourable under standard conditions at 298.15 K? Table 11.3 Thermodynamic data at 298.15 K for scandium (Sc)
90
Now that our introduction to some of the basic ideas of chemical thermodynamics is complete, we shall remind you of some of the problems and questions that we elicited earlier in Sections 2-4 of this Book. Sections 2.2-2.4 discussed and observed the reactions of metals with H+(aq), with the aqueous ions of other metals, with oxygen and with the halogens; Section 3 briefly discussed the extraction of metals from their ores. In Section 4, we elicited a plausible hypothesis from these experiments and observations. It seemed as if it might be possible to arrange metals in order of their reactivity or their tendency to be oxidized. This order looked as though it might possibly remain the same even when we changed the oxidizing agent. In Section 4.1 we looked at this hypothesis more critically. Firstly, we noted that none of the experiments or observations on which the hypothesis was based was quantitative. Secondly, we scrutinized the language in which the hypothesis was expressed, in particular the concepts of reactivity and tendency to be oxidized; neither concept took account of the crucial distinction between the rate of a reaction and its equilibrium position. In Section 4.1, we remarked that both these criticisms could be alleviated by comparing equilibrium constants for the reactions of metals with a particular oxidizing agent. We have now provided you with the means to do this. In Section 11.I , we showed that for a reaction at constant temperature, the standard Gibbs function change, AG:, is directly related to the equilibrium constant. Values of AGE, therefore, provide us with a way of comparing equilibrium positions for the reactions of metals with particular oxidizing agents, and in the next Section, we shall illustrate the point for the case in which the oxidizing agent is H+(aq). There we embark on an approach that will eventually provide answers to questions that lie at the heart of the problem discussed in Sections 2-4. These questions are: 1
How can we compare values of AGE for the reactions of metals with a particular oxidizing agent?
2
What is meant by the ‘reactivity’ of a metal, and how is it related to the value of AGE and to the rate of the reaction concerned? Do values of AG: for the reactions of metals with one particular oxidizing agent remain in the same order when the oxidizing agent is changed? What atomic properties cause one metal, such as sodium, to be a better reducing agent than another, such as silver (Figure 12.1)?
3
4
Figure 12.1 The strength of sodium (a) as a reducing agent becomes apparent when water is dropped on it; the water is reduced with such vigour that the remaining sodium catches fire. The weakness of silver (b) as a reducing agent is revealed by its endurance in works of art such as the Entemena vase of ancient Sumeria, dated 2500 :BC.
91
In Sections 2-4 we asked whether we could grade metals in the order of their tendency to be oxidized, and in Section 2.2 we were especially interested in their oxidation to aqueous ions in reactions such as Mg(s) + 2H+(aq) = Mg2+(aq)+ H2(g>
(2.2)
In Section 4.1, we set out with the intention of comparing different metals in terms of the position of equilibrium in instances such as Reaction 2.2. Several problems had to be overcome. Firstly, the equilibrium position in Reaction 2.2 lies well over on the right-hand side of the equation (Figure 12.2 left), and it is impossible to obtain a value of the equilibrium constant by measuring the concentrations of the species in the equation at equilibrium. Secondly, we saw in Section 2.2 that there were reactions similar to Reaction 2.2 that did not occur, and it seemed impossible to compare the equilibrium positions for these reactions; for example, silver and dilute acid do not react (Figure 12.2 right) (2.10) The standard Gibbs function change, AGE,goes a long way towards providing a solution to these two difficulties. From Equations 11.7 and 11.8, we know that, at : G A the larger is the equilibrium conconstant temperature, the more negative is , stant of a reaction. What is more, the values of AGE do not necessarily have to be
Figure 12.2 Magnesium (left) reacts vigorously with dilute acid (Reaction 2.2); silver (right) does not. 92
determined by studying the reaction itself; they can be calculated from the results of experiments on the individual reactants and products (usually obtainable from tables of published data). Values of AGg for most reactions are calculated from standard enthalpy and entropy data, because this method has the widest application. For Reaction 2.2 at 298.15 K: AH: = -466.9 kJ mol-'; AS: What is the value of AG: at 298.15 K?
= -40.1 J K-l mol-'.
The answer is -454.9 kJ mol-l. The figure is obtained by substituting appropriate values in the equation
AGE= AH:
-
(11.4)
TAS:
(Remember that T = 298.15 K and that AS: not kJ K-' mol-l).
was quoted in the unit J K-' mol-l,
Repeat the calculation for Equation 2.10, given that AH: and AS: = 95.5 JK-I. This time, AG:
= 105.6kJ mol-'
= +77.1 kJ mol-'.
Note how the figures concur with the observations made on Reactions 2.2 and 2.10 in Section 2.2. The value for the magnesium reaction is large and negative, so, thermodynamically, the reaction is very favourable. This is in accord with our observation that magnesium dissolves in acids to give hydrogen gas and Mg2+(aq) ions. The value for the silver reaction is large and positive, so the reaction is thermodynamically unfavourable and the equilibrium position lies well over on the lefthand side. This is in accord with our observation that silver does not dissolve in dilute acids to give hydrogen gas and Ag+(aq) ions. Let us now move on to the question of whether the two values we have given for AG: for Reactions 2.2 and 2.10 (-454.9 kJ mol-' and +77.1 kJ mol-', respectively) provide a quantitative basis for comparing positions of equilibrium. Each value refers to a particular amount of material -for one mole of reaction for the particular Equations 2.2 and 2.10. The most useful comparison is procured by a careful choice of the amounts of material to which the values of AGg refer. It can be seen that in Equation 2.2, one mole of metal reacts with two moles of hydrogen ions to produce one mole of hydrogen gas. As we shall show more clearly in a moment, the values of AG: are most usefully compared when they refer to reactions in which the metals react with the same number of moles of aqueous hydrogen ions to produce the same number of moles of hydrogen gas. Convert Reaction 2.2 (and its AG: value) to one in which the number of moles of hydrogen ions is the same as in Reaction 2.10. One mole of hydrogen ions is needed on the left-hand side and half a mole of hydrogen gas on the right, so the magnesium reaction and its AG: value must be halved: 1
1
1
ZMg(s) + H+(aq) = 5Mg2+(aq) + ZH2(g); AG:
= -227.5 kJ mol-l
(12.
93
Subtract Equation 2.10 and its AG: value.
value from Equation 12.1 and its AGE
All the hydrogen disappears, and you should get 1
?Mg(s)
+ Ag+(aq) = kMg2+(aq)+ Ag(s); AG:
= -304.6kJm01-~
(12.2)
How is this last value related to the observations in Activity 2.3? The equation describes the reaction of magnesium with silver ions to give silver metal and magnesium ions; the large negative value of AG: shows that equilibrium lies well over to the right. These conclusions are consistent with Activity 2.3, where you saw that the reaction happened. Let us now expand this kind of operation. We can write down a series of equations for the various reactions between metals and hydrogen ions in which one mole of H+(aq) appears on the left and half a mole of hydrogen on the right. These equations will be of the general kind: 1 1 1 -M(s) + H+(aq) = -Mn+(aq) + ,H2(g) (12.3) n
where n is the number of positive charges on the metal cation. We can then use experimental thermodynamic data to obtain the values of AGE for each of these reactions, and then arrange the reactions in a table with the most negative values of AG: at the top and the most positive at the bottom. This has been done in Table 12.1. If you again thnk carefully through the example of magnesium and silver, you will see that Table 12.1 has been arranged so that at 298.15 K any metal is thermodynamically capable of reducing the ions of another metal below it in the table to the metal itself, the first metal being simultaneously converted into its ions. This follows because AG: for the reaction must be negative at 298.15 K. To convince yourselfof this, try the following Question.
Use Table 12.1 to calculate values of AGE for the following reactions: 1 1 (i) ZZn(s) + Ag+(aq) = ?Zn2+(aq) + Ag(s) (ii) Cu(s) + Mg2+(aq) = Cu2+(aq)+ Mg(s) (iii)Fe(s) + 2Ag+(aq) = Fe2+(aq) + 2Ag(s) (iv) 3Hg(l) + 2A13+(aq)= 3Hg2+(aq) + 2Al(s) The series of reactions for the metals in Table 12.1 is often called the electrochemical series. It gives us a means of comparing equilibrium constants for the reactions of metals with the oxidizing agent H+(aq). If you read other textbooks, you will often find it graded in terms of another quantity called 'the redox potential, p', rather than in values of AG: . You will meet this alternative later in Section 21.
The series in Table 12.1 supplies the answer to the first of our introductory questions in the particular case where the oxidizing agent is H+(aq): from thermodynamic considerations alone, the most readily oxidized metals are at the top of the series, and the least readily oxidized are at the bottom. However, as you will see in the next Section, thermodynamics alone does not determine whether a reaction occurs or not, and we must explore this point further before we tackle our second introductory question on the meaning of reactivity. At this stage, you must be able to manipulate 94
Table 12.1 Standard Gibbs function changes for the reactions of metals with aqueous hydrogen ions at 298.15 K
for the aqueous cations recorded in the Data the values of AG?, AH? and Book. The procedure was covered in Question 11.5. Remember that the values of AGF and AH? are the values of AG: and AH: for reactions of the type: M(s)
+ nH+(aq) = Mn+(aq)+ ;H2(g)
(12.4) 95
You have reached an excellent level of performance if you can do Question 12.1 above and Questions 12.2 and 12.3 below!
Metals that form monatomic aqueous cations can be arranged in order of the values of AG: for the reaction 1 + H+(aq) = -M"+(aq) 1 + ;H,(g) -M(s) (12.3) The tendency for the metal to be oxidized increases as AGE becomes more negative.
This question is concerned with Table 12.2. Lanthanum (Figure 12.3a) is the first member of the lanthanide series, and plutonium (Figure 12.3b and c) is an actinide metal whose isotopes are used in nuclear reactors and nuclear weapons. Both metals dissolve in dilute acids to give hydrogen and M3+ions. Table 12.2 Thermodynamic data at 298.15 K for lanthanum and plutonium
An investigator determines the values of AH: and AS: reactions La(s) + 3H+(aq) = La3+(aq) + +H2(g) Pu(s)
+ 3H+(aq) = Pu3+(aq) +
at 298.15 K for the
;H2(g)
For the lanthanum reaction, the values found are AH: = -709.0 kJ mol-I and A S : = -84.1 J K-l mol-' For the plutonium reaction, the values found are AH: = -593.0 kJ rno1-I and AS: = -25.9 J K-1 mol-I Fill in the six blank entries in Table 12.2.
What positions would lanthanum and plutonium occupy in Table 12.1?
96
(12.5) (12.6)
Figure 12.3 Samples of lanthanum, (a), and plutonium, (b), the two metals that are the subject of Questions 12.2 and 12.3. The photograph of plutonium has been taken by the light produced by its radioactive decay. In (c), the lower blue layer shows the colour of the aqueous plutonium ion, Pu3+,which also features. It is undergoing separation from a yellow solution of uranium in the upper (kerosene) layer.
97
You know that two separate conditions must be fulfilled before a reaction can occur: first, the equilibrium position must favour the reaction, and second, the reaction must have a measurable rate. The thermodynamics introduced in Sections 5-1 1 allows us to restate the first of these two conditions: we can now say that a reaction will occur only if AG: is negative, and if the reaction has a measurable rate. Clearly, if some reaction does not occur, it could be due to a failure on either of these two fronts. Firstly, AGE could be positive: in such a case, we say that the reaction is thermodynamically unfavourable and that the reactants are thermodynamically stable with respect to the products. Alternatively, AGE could be negative. If this is so, then the reaction is thermodynamically favourable, and the absence of an observable reaction must be due to an immeasurably slow rate. In such a case, the reactants are said to be kinetically stable with respect to the reaction in question. Faced with a reaction that does not happen, you should now be able to decide which of these two kinds of stability is responsible. To illustrate this, we shall use the reactions 1 p 2 @
Cu(s)
= NH3(g)
(13.1)
+ 2H+(aq) = Cu2+(aq) + H2(g)
(2.9)
+
3 3 2 w
At 298.15 K, neither of these two reactions occurs. When nitrogen and hydrogen are mixed at this temperature, nothing happens: no ammonia is formed. Likewise, as you saw in Activity 2.1, copper does not react with dilute hydrochloric acid. Use the Data Book and Table 12.1 to find out whether the reactants in Equations 13.1 and 2.9 are thermodynamically or kinetically stable with respect to the products. In Reaction 13.1, ammonia is formed from its constituent elements, and AGE is equal to AGF(NH3, g). From the Data Book, AGF(NH3, g) = -16.5 kJmol-l. Because AGE is negative, the equilibrium position in Equation 13.1 lies well over to the right. This means that the reaction does not occur because it is very slow: at 298.15 K, nitrogen and hydrogen are kinetically stable with respect to ammonia. For Reaction 2.9, we see from Table 12.1 that AG? = 2 x 32.7 kJ mol-1 = 65.4k.Jm01-~. AGE is positive, so equilibrium lies to the left of Equation 2.9: at 298.15 K, copper and H+(aq)are thermodynamically stable with respect to Cu2+(aq)and hydrogen gas. The analysis of stability in terms of thermodynamics and kinetics is most important. When we want to find out why a reaction does not occur, this analysis is the first step in any rigorous argument. The procedure used for Reactions 13.1 and 2.9 is quite general: it is shown diagrammatically in Figure 13.1.
98
Figure 13.1 Analysis of the reasons why a reaction fails to occur.
The thermodynamics of a reaction depends only on the initial and final states of the reactants and products: thermodynamics tells us nothing about the intermediate steps, the processes that occur as the reactants are converted into the products. These processes, which constitute the mechanism of the reaction, are the province of reaction kinetics*. In the case of Reaction 13.1, for example, the mechanism is the process by which the very strong N=N and H-H bonds in the N2 and H2 molecules are broken, and N-H bonds are formed. Somewhere in this mechanism lies the reason for the kinetic stability of a mixture of nitrogen and hydrogen at 298.15 K. When, as in the case of Reaction 13.1, a thermodynamically favourable reaction is immeasurably slow, we can take steps to increase the rate. What steps are taken in the case of Reaction 13.1? Reaction 13.1 is the Haber process for the preparation of ammonia. As you know, the reaction can be speeded up by using a catalyst -in this case iron and by raising the temperature to about 750 K. The introduction of a catalyst and an increase in temperature are quite common methods of speeding up a reaction. Other measures may be appropriate in particular cases. Consider, for example, the statue of Eros (Figure 13.2). This was one of the earliest pieces of aluminium casting: it has been standing in Piccadilly Circus since 1893. Now, AG7(Al2O3, s) = -1 582.3 kJmol-'. Is the survival of Eros due to its kinetic or thermodynamic stability? Kinetic stability: the value of AGF shows that the combination of the metal with atmospheric oxygen is thermodynamically favourable. The lack of corrosion is caused by the slow rate of reaction.
* Chemical kinetics is the subject area of Chemical Kinetics and Mechanism'.
Figure 13.2 The statue of Eros in Piccadilly Circus is one of the earliest examples of aluminium casting. 99
This slow rate has an interesting cause: when a clean piece of aluminium foil is exposed to oxygen, a thin film of A1203,about m thick, is quickly formed on the metal surface. This film prevents further contact between the metal and oxygen so that the reaction then ceases: here the mechanism of the reaction incorporates a step that stops the reaction dead! One way of speeding up the reaction is to bring the reactants into more intimate contact by grinding the aluminium into a very fine powder. This increases the surface to volume ratio so much that oxide formation does not stop with a surface layer: very fine aluminium powders can actually explode spontaneously in air. A more subtle method is shown in Activity 4.1 (on the CD-ROM associated with this Book): a little mercury or mercury(I1) chloride will destroy the coherence of the oxide film, and the underlying aluminium dissolves in the mercury to form an alloy called an amalgam. The aluminium in the amalgam will then combine quickly with atmospheric oxygen to form white flakes of oxide (Figure 13.3).
Figure 13.3 Unlike untreated aluminium (left), aluminium whose surface has been briefly amalgamated by contact with mercury dichloride solution reacts vigorously with atmospheric oxygen or water.
The factors that affect the kinetic stability of aluminium at 298.15 K are clearly very different from those responsible for the kinetic stability of a mixture of nitrogen and hydrogen at the same temperature. They show how wide-ranging in scope such factors can be: kinetic stability can be attributed to anything that prevents a thermodynamicallyfavourable reactionfrom taking place.
100
1
2
3
If AG: for a reaction is positive at any temperature, the reactants are thermodynamically stable with respect to the products: the reaction will never occur at that temperature. If AGE for a reaction is negative, the reaction may occur or it may not. It will occur if the rate of reaction is sufficiently high. It will not occur if the rate of reaction is immeasurably slow -that is, if the reactants are kinetically stable with respect to the products. An immeasurably slow but thermodynamically favourable reaction can possibly be speeded up by finding a catalyst, or by bringing the reactants into more intimate contact, or by increasing the temperature.
Ignoring the possibility of slow rates of reaction, which of the four reactions in Question 12.1 would you expect to occur?
In Activities 2.3 and 4.1, you saw that the following reactions did not occur: Cu(s) + Mg2+(aq)= Cu2+(aq)+ Mg(s) 2Al(s) + 3Cu2+(aq) = 2A13+(aq)+ 3Cu(s) Decide in each case whether the lack of reaction is due to kinetic or thermodynamic stability.
In proving that a stable chemical system is kinetically rather than thermodynamically stable, one must find a reaction that the system might undergo and show that it has a negative value of AG:. Do this for the following systems in which no reaction occurs at 298.15 K: (a) a mixture of diamonds and oxygen at 298.15 K; (b) a mixture of magnesium ribbon and solid zinc chloride (ZnC12) at 298.15 K.
101
We have now arrived at the point where we can begin to answer the second of the three introductory questions in Section 12. This was concerned with the meaning of reactivity. In Sections 2-4, we used merely visual observation of reactions to grade metals in order of their reactivity or tendency to be oxidized. Let us look at a typical example of the style of reasoning that we used: ‘Suppose we take three beakers of dilute hydrochloric acid, and drop a piece of silver into the first, lithiuh into the second and potassium into the third. In the first case, nothing happens. In the second, the lithium floats steadily around on the surface evolving hydrogen. In the third case, the potassium streaks around the acid surface, and the evolved hydrogen catches fire (see Figure 14.1). Clearly potassium is more reactive than lithium towards dilute acid, and lithium is more reactive than silver.’ This kind of reasoning is compelling because it is directly derived from the full drama of what we see. We shall now show how it conflicts with the logic of Section 13. Table 14.1 lists values of AG: and AH: for the appropriate reactions of the three metals. Table 14.1 Thermodynamic data for the reactions of lithium, potassium and silver with dilute acid
Figure 14.1
The reactions of (a) silver, (b) lithium and (c) potassium with dilute acid.
Are the differences in reactivity attributable to a kinetic or thermodynamic effect? The silver reaction is distinguished from the other two by a positive value of AG: : it is thermodynamically unfavourable. The difference between silver on the one hand and lithium or potassium on the other is due to thermodynamic effects. But the difference between lithium and potassium is much more complicated.
102
The values of AG: show that equilibrium is further to the right for lithium, and the values of AH: show that more heat is evolved per mole of metal in the lithium reaction. The reason why potassium is more reactive must be that its reaction with hydrogen ions is faster. Heat is generated quickly in the potassium reaction, so there is a localized increase of temperature around the dissolving metal, which causes the hydrogen to catch fire. The symptoms by which we recognize reactivity, such as marked rise in temperature and violent agitation of the reactants and products, arise from a peculiar combination of kinetic and thermodynamic effects. We can distinguish two quite basic requirements, and a third which is less important. 1 One obvious requirement that must be fulfilled before a chemical system can be described as reactive is that the reaction should be very fast -a question of kinetics. 2 Equally obvious is the requirement that the reaction should be thermodynamically favourable -a question of thermodynamics that is contained in the condition that AGE must be negative. 3 Visual impressions of reactivity are enhanced if AH: is large and negative, because if the initial reaction is fast, heat will be evolved quickly, leading to a localized rise in temperature. This rise in temperature increases the rate still further, and the symptoms of reactivity become even more marked. To summarize, the observed ‘reactivity’ is an ill-defined union of thermodynamic and kinetic effects. ‘Reactive’ materials usually display their reactivity in reactions that not only have large equilibrium constants and large negative values of AH:, but also proceed rapidly. Two conclusions follow from this. Firstly, when you say that a substance is reactive, try to specify those reactions through which its reactivity is expressed. In other words, compare the reactivity of reactions rather than that of individual substances. When you have done this, analyse the behaviour in terms of thermodynamics and kinetics. Secondly, since quantitative scientific accuracy demands that we separate kinetic and thermodynamic effects, we cannot use ‘reactivity’ to compare the reactions of metals. We must make either kinetic or thermodynamic comparisons. Because we can compare the thermodynamics but not the rates in cases where reactions do not occur, we shall in fact use thermodynamics to make the comparison. It is on these terms that we shall tackle the third of our three introductory questions.
103
In this Section, we shall use thermodynamics to explore the extent to which we can grade metals in order of their ease of oxidation. The experiments and observations of Sections 2-4 suggested that it might be possible to produce a single grading of metals arranged in order of their tendency to be oxidized by various oxidizing agents. As we said at the time, this idea was prominent in the nineteenth century, and it led to quite detailed but essentially qualitative gradings, of which the following example is typical:
Cs > Rb > K > Na > Li > Ba > Sr > Ca > Mg > A1 > Zn > Fe > Sn > Pb > H > Cu > Hg > Ag > Au Now, in Section 12.1 we used thermodynamic data to set up a quantitative grading of metals in order of their thermodynamic willingness to be oxidized by aqueous hydrogen ions. Compare the grading given immediately above with that in Table 12.1. Do the two correspond (a) in broad outline, (b) in detail? They certainly correspond in broad outline. For example, in each case the alkali and alkaline earth metals come first, whereas copper, silver and gold are last. Detailed study shows that from magnesium onwards, the two series are almost identical, but there are serious discrepancies among the first eight metals. Compare, for example, the positions of lithium and sodium in the two series. Similarities of this kind suggest one of two possibilities: firstly, there might be a unique series (but either one or both of the two that we have compared gives untrustworthy results for the first eight metals); secondly, there is no unique series. We can resolve these alternatives by setting up another table in which the metals are arranged in order of their thermodynamic willingness to be oxidized by a different oxidizing agent. In Table 12.1, we used aqueous hydrogen ions as the oxidizing agent, and wrote down a series of equations in which the same amount of H+(aq) (one mole) appeared on the left-hand side. On subtraction of the equations, hydrogen ions and hydrogen gas disappeared. But from a practical standpoint, the most important oxidizing agent to which metals are vulnerable is the molecular oxygen in the air. The oxidizing power of molecular oxygen, often augmented by the presence of water, is the source of metal corrosion, an immensely costly and wasteful process. In Table 15.1, we have used molecular oxygen as our oxidizing agent and written down the equations with the same amount of oxygen (half a mole) on the left-hand side. Again, on subtraction of the equations, oxygen gas disappears. The equations have been arranged, as in Table 12.1, with the most negative values of AGE at the top. (The values are taken from the Data Book.)
104
Table 15.1 Values of AGE for the reactions of different metals with molecular oxygen at 298.15 K
You can, of course, manipulate the equations and AGE values in Table 15.1 in exactly the same way as you did with the ones in Table 12.1. This exercise forms part of Question 15.1. However, here we concentrate on the comparison between the order of the metals in Tables 12.1 and 15.1. Compare the order of the metals in Tables 12.1 and 15.1. Do the two series correspond (a) in broad outline, (b) in detail? The similarities are not as close as last time, but they are still apparent. Again the alkali and alkaline earth metals come at the top, whereas copper, silver and gold are at the bottom. This time, the detailed differences are more striking: compare, for example, the positions of magnesium and aluminium. Which of the two possibilities mentioned earlier in this Section is closer to the truth? The second; the differences between the two sequences in Tables 12.1 and 15.1 suggest that the concept of a unique series is an illusion. No single series can express the thermodynamic tendency of metals to be oxidized by different oxidizing agents, and the goal of a unique series, whose possible existence was suggested in Sections 2.3 and 4, is unattainable. However, just as important is the fact that the series do have overall similarities: the alkali and alkaline earth metals are near the top, whereas silver and gold are at the bottom. We shall return to this point later in the Book.
We start this Section with a question: For how many of the metals in Table 15.1 is the value of AG negative? For all except gold. This is very remarkable. Only for gold is AG?(Au2O3, s) positive. In all the other cases, oxidation of the metal by atmospheric oxygen is thermodynamically favourable at 25 "C. Because we rely so heavily on metals that exist in contact with atmospheric oxygen, our civilization depends on the fact that the oxidation of all metals in common use is slow. Only kinetic barriers stand in the way of the structural collapse of many of life's essentials, including cutlery, the car and the jumbo jet. Another feature of Table 15.1 is that for most of the metals in it, the values of AG: are very negative. Now from Equation 11.4: (11.4) = AH: - TAS:
AGE
So the negative values arise either from very negative values of AH: or very positive figures for AS:. We shall have more to say about this matter in Section 16. However, the essential point is that very negative values of AH: are usually responsible. Consequently, the oxidation of many of the metals in Table 15.1 is associated with the evolution of large amounts of heat. In Section 14 we noted that this makes it likely that, once started, the reaction will proceed at an increasing and perhaps catastrophic rate. The case of aluminium shows how tenacious oxide films are often the reason why vigorous reaction does not occur. In Section 14, you saw that corrosion is rapid if the film is disrupted by amalgamation. Small particle sizes, at which the surface-tovolume ratio is much larger than for bulk aluminium, also threaten the coherence of the film. In workplaces, where aluminium powder is handled, or aluminium dust circulates, a small spark can ignite the metal and cause accidents. Not surprisingly then, aluminium powder is used in fireworks and flares to produce an intense white light. The next Section describes how a similar reaction has been invoked in an explanation of the strange phenomenon of ball lightning.
15.1 .1
Ball lightning
Everyone is familiar with the common type of lightning, shown in Figure 15. l a as a jagged discharge or 'bolt'. Ball lightning, as described in Box 15.1 by the American physicist, Graham Hubler, is very different. It consists of a glowing sphere of light, about the size of a grapefruit, which usually appears out of thin air during thundery weather. The spheres float close to the ground, drifting slowly for about 20 seconds before expiring. Recorded sightings of ball lightning go back to the Middle Ages, although its occurrence is very rarely observed. A few photographs exist (Figure 15.lb), but as yet [2002], there are no known videotapes. Until recently, there was no convincing scientific explanation; it sounds, in fact, like the stuff of which UFOs are made.
106
107
A recent explanation of ball lightning depends on properties of the semi-metal silicon. The combination of silicon bound to oxygen is the principal component of soils and rocks; sand, for example, is mostly the oxide of silicon, Si02.The reaction between silicon and oxygen is thermodynamically very favourable: 1
ySi(s)
+
1
1
T02(g) = TSi02(s); AGE = -428.3 kJmol-’
(15.1)
Where does this value place silicon in Table 15.1? Beneath barium and above sodium. The thermodynamic favourability of the oxidation of silicon is therefore comparable with that of highly reactive alkali and alkaline earth metals. It is also associated with the evolution of much heat: for Reaction 15.1, AH: = - 455.5 kJ mol-l. So if hot silicon is created in a finely divided state, we can expect a vigorous exothermic reaction in the presence of oxygen. In 2000, two chemical engineers in New Zealand, John Abrahamson and James Dinniss, used this idea to produce a new and promising explanation of ball lightning. A conventional lightning strike on soil can generate temperatures as high as 3 000 K. Silicates or silica could then react with carbon from the organic matter in the soil to produce a vapour containing silicon atoms. This vapour is blown into the air by the shock wave of the lightning strike. As the silicon atoms cool, they combine to form a buoyant fluff ball composed of chains of tiny particles of solid silicon. This fluff ball burns steadily in air as the finely divided silicon combines with oxygen to regenerate Si02. When the silicon has been consumed, the glowing sphere disappears. Abrahamson and Dinniss exposed soil to a lightning-like discharge in the laboratory. Although they did not see ball lightning, they did find the required chains of tiny silicon particles in the air space close to the strike. This is promising. Experiments are under way to find the soil and discharge conditions that might generate ball lightning.
1
2
‘Reactive’ materials usually reveal their reactivity by reactions that not only have large negative values of AH: and AG:, but are also fast. Thus, ‘reactivity’, as usually understood, is a combination of thermodynamic and kinetic factors. There is no single order of metals that expresses their thermodynamic tendency to be oxidized; different oxidizing agents yield different orders. However, some metals, such as the alkali metals, are invariably more readily oxidized than others, such as copper, silver and gold.
Use the data in Table 15.1 to calculate values of AG: reactions at 298.15 K: (i) 2Al(s) + 3PbO(s) = A1203(s) + 3Pb(s) (ii) Cu(s) + Li20(s) = CuO(s) + 2Li(s)
108
for the following
Use information from the Data Book to construct a table similar to Tables 12.1 and 15.1, representing the thermodynamic tendency of the metals Ag, Al, Ca, Cu, Hg, Na and Zn to be oxidized by chlorine at 298.15 K. Does the table support the conclusions reached in Section 15?
Reactions obtained by subtracting equations in Table 12.1, and predicted to be thermodynamically favourable, very often occur at 298.15 K. By contrast, reactions obtained by subtracting equations in Table 15.1 and predicted to be thermodynamically favourable, very often do not occur at 298.15 K. Suggest a reason for this difference.
109
Before moving on to the next stage of our investigation of the redox reactions of metals, we must look a little more closely at the enthalpy and entropy components of the Gibbs function. The important equation is (11.4) A G =~ AH: - T A S ~ A reaction is thermodynamically favourable if AG: is negative. According to Equation 11.4, this situation is furthered by negative values of the standard molar enthalpy change, AH:, and positive values of the standard molar entropy change, AS:. Conversely, positive values of AH: and negative values of AS: tend to make reactions thermodynamically unfavourable. The sign of AG: is determined by the balance between the enthalpy and entropy terms on the right-hand side of Equation 11.4.
By surveying the thermodynamics of a wide variety of chemical reactions at normal temperature, one finds that in most cases, the enthalpy term is more influential than the entropy term in determining the sign of AG:. Let us begin by looking at the thermodynamics of some reactions from Table 15.1. Consider the reaction zn(s) + &(g)
(16.1)
= ZnO(s)
at 298.15 K. Here, using the Data Book,
AS:
= P ( z n 0 , s) - P ( z n , s) -
;s~(o,,g>
= (43.6 - 41.6 - 102.6) JK-1 mol-1 = -100.6 J K-1 mol-1
Now the Data Book also tells us that AH: = -348.3 kJ mol-l. Thus AG: = -348.3 kJmol-1 - (298.15 x -100.6) Jmol-l = -348.3 kJ mol-1 + 30.0 kJ mol-1 = -3 18.3 kJ mol-1
Notice that in this calculation, the numerical value of the TAS: term (30.0kJmol-1) is only about 9% of the numerical value of the AH: term (348.3 kJ mol-1). We show this diagrammatically on the left of Figure 16.1. Because TASZ is small compared with AH:, AG* does not differ greatly from AH:. For most reactions at room temperature, AHEmakes a much larger contribution to AGE than does T AS:. Indeed, this fact explains the plausibility of Thomsen’s hypothesis, which you met earlier in the Book (see, for example, Section 6). In Equation 11.4 at 298.15 K, how many J K-l mol-1 in the AS: equivalent to 1 kJmol-1 in the AH: term?
term are
If T A S E is to make a 1 kJ mol-1 contribution to AG: at 298.15 K, AS: must be 1 000/298.15 = 3.35 J K-l mol-l. Thus, to the nearest whole number, an entropy change of 3 J K-l mol-1 is equivalent to only 1 kJ mol-l.
II 0
Because TAS: is often small compared with AH:, we shall occasionally ignore However, we shall most it altogether and assume that AGE is equal to AH.: frequently ignore the TASZ term when we are comparing the values of AG: for similar types of reaction. To demonstrate this, we can compare the formation of zinc oxide, ZnO, with that of magnesium oxide, MgO. For the reaction
A S :
= Se(Mg0, S ) - Se(Mg, S) - iSe(O2, g) = (26.9 - 32.7 - 102.6) J K-l mol-1 = -108.4 J K-' mol-'
Notice that the value of A,!?: for Reaction 2.5 differs by only about 8 J K-' mol-1 from the value for Reaction 16.1 (-108.4 J K-' mol-1 and -100.6 J K-l mol-l, respectively). As we move from the zinc to the magnesium reaction, the molar entropy of the metal decreases substantially, but so does the value for the oxide. Hence the change in the difference between them is relatively small (about 8 J K-l mol-l). The large value of i p ( 0 2 , g) is common to both reactions, so it does not affect the difference between the values of AS:. Using the Data Book, we obtain for the magnesium reaction at 298.15 K AH: = -601.7 kJ mol-1 and
AG:
= -601.7 kJ mol-* - (298.15 x -108.4) J mol-1 = -569.4 kJ mol-l
Figure 16.1 Relative magnitudes of AG:, AH: and T A S Z for two reactions at 298.15 K: reaction of zinc with oxygen (left); reaction of magnesium with oxygen (right).
111
We can compare the two sets of calculations diagrammatically, as in Figure 16.1, or numerically. You will notice that for both reactions, the TAS: term is small compared with AH:. But notice too, that the difference between the values of AG: (25 1.1kJ mol-I) is almost entirely due to the difference in the values of AH: (253.4 kJ mol-l). This is because the two values of AS: are so similar. The result is that when we compare two values of AG: for similar types of reaction, the error incurred by ignoring the AS: term and simply comparing the AH: values is usually very small. Before closing this Section, we shall say a little more about the meaning of the expression, ‘similar types of reaction’. In the case of Reactions 16.1 and 2.5, it means that the equations differ only because one element (magnesium) has been substituted for another (zinc). We shall call reactions that are related in this way analogous reactions. Particularly important is the fact that the substitution does not involve any change in the numbers that are in, or that precede, chemical formulae, or in the physical states of reactants and products. Thus, solid zinc and solid zinc oxide are replaced by solid magnesium and solid magnesium oxide. The importance of this proviso should become more obvious when you have read the next Section.
In most instances, the form of a chemical equation reveals nothing about the sign or magnitude of AG:, AH: or AS:. However, from Section 10.6 you know that, when the reaction involves only pure substances, some of which are gases, the sign of AS: can usually be predicted with reasonable confidence. Consider, for example, Reactions 16.1 and 2.5. These have similar large negative values of AS:. What feature of the equation leads us to expect this? From Section 10.5, you know that, in general, gases have much larger molar entropies than solids or liquids. Now in the solid-gas Reactions 16.1 and 2.5, there is half a mole of oxygen gas on the left-hand side of the equations, but no moles of gas on the right. The entropy of the reactants is therefore much greater than that of the products. Thus, one obvious reason why the two analogous reactions have similar large negative entropy changes is that they both involve a decrease of half a mole of the same gas per mole of reaction. Conversely, a reaction involving solids and gases which leads to an increase in the number of moles of gas will have a large positive value of AS: (Figure 16.2).
112
Figure 16.2 (a) The combustion of magnesium consumes oxygen gas and has a large negative entropy change; (b) the explosion of solid nitrocellullose produces gases, and has a large positive entropy change.
For the majority of chemical reactions at 298.15 K, AH: in the equation AG: = AH: - T A S ~ (11.4) makes a bigger contribution to AG: than does -TAS:. For a series of analogous reactions at 298.15 K, the variations in AG: are determined almost entirely by those in AH: because the TAS: values are similar. Reactions involving solids and gases have large positive values of AS: when there is an increase in the number of moles of gas, and large negative values of AS: when there is a decrease in the number of moles of gas.
In one of the following three pairs of reactions, the two reactions have very similar values of AS:; which pair? (i)
+
CuO(s) = Cu(s)
;02(g)
HgO(s) = Hg(g) + ;02(g) (ii) MgC03(s> = M g W ) + C02(g) CaC03(s) = CaO(s) (iii) Sn02(s) = Sn(s) SnO(s) = Sn(s)
+ Co2(g)
+ 02(g)
+
1
,o2(g)
From reactions (i)-(v) below, select those that (a) have large negative entropy changes, (b) have large positive entropy changes, (c) have small entropy changes that may be either positive or negative.
113
In Sections 2.4 and 3, we pointed out one of the most striking features of the chemical behaviour of metals: a particular metal often responds similarly towards quite different oxidizing agents. We indicated a relation between this observation and the details of metal extraction from ores. Thus, gold reacts very little, if at all, with the various corrosive substances found in the lithosphere and hydrosphere. Consequently, it is found naturally in the uncombined state. By contrast, the alkali and alkaline earth metals react vigorously with water, oxygen, chlorine or hydrogen chloride, and are never found free in nature. They occur as aqueous ions, chlorides, etc., which must be reduced if the metal is to be isolated. We distinguished three classes of metals: 1 ‘Noble’ metals, which are either found free or can be obtained by gently heating their ores in air (Au, Ag, Hg; see Figure 17.1). 2 Metals that are obtained by heating their ores in air to form oxides, which can then be reduced by heating with carbon (Pb, Sn, Fe, Zn; see Figure 17.2).
Figure 17.2 Molten lead being produced by heating carbon with orange lead monoxide (litharge or PbO).
3
Figure 17.1 The nobility of silver illustrated by the endurance of one of the most famous of all silver coins. The Athenian silver tetradrachm of the fifth century BC was the monetary symbol of the economic power of the Athenian empire; on one side is the ‘almond-eyed’ goddess Athena, and on the other, her symbol, the owl.
Metals that are not easily obtained by carbon reduction and were isolated in relatively recent times, either with the help of electrolytic methods, or by using metals in the same class as reducing agents (alkali and alkaline earth metals, aluminium; see Figure 17.3).
Figure 17.3 Caesium is made by heating its molten chloride with calcium, and stored in sealed ampoules. The solid is silvery, but if an ampoule is held in the hand, the caesium melts, and becomes a golden liquid.
Although rates of reaction are important in metal extraction, it is still very helpful to view this classification from a thermodynamic standpoint. We shall now do this.
114
Until now, we have handled AG: values at only one temperature: 298.15 K. But many reactions and industrial processes, including metal extraction, must be carried out at higher temperatures. Sometimes there is a need to speed up reactions that are kinetically unfavourable; at other times the intention is to improve the yield by increasing the equilibrium constant. But whatever the reason for the heating process, the values of the equilibrium constant and A G : are different at the higher temperature. This brings us up against a question of the greatest importance: how does AGE change with temperature? We know that for any reaction at any temperature, = AH: - TAS:
AGE
Tables like those in the Data Book enable us to calculate , : G A reactions at 298.15 K.
(11.4)
AH: and AS:
for
Suppose now that the temperature is increased substantially, and that the physical states of the reactants and products remain what they were at 298.15 K; that is, no reactant or product melts or boils within the range of temperature increase. It is then found experimentally that the values of AH: and A S : change relatively little. As an illustration of this fact, Figure 17.4 shows plots of , : G A against temperature for the reaction MgW +
;om
= MgO(s)
AH: and A S : (2-5)
in the range 250-900 K. Magnesium and magnesium oxide both melt above the top of this temperature range, but within it no phase changes occur. The values S : are remarkably constant throughout the range, but AGE of AH: and A changes by quite large amounts. Many other examples of the same effect could be given. We shall therefore use the following approximation for all chemical reactions: provided that no reactant or S : do not change as the temperature is raised product melts or boils, AH: and A above 298.15 K.
Figure 17.4 Variation with temperature of AGE, and for Reaction 2.5. AG; and AH: values relate to the left-hand axis, and : S A values to the right-hand axis. 115
It follows that in Equation 11.4 we can use the values of AH: and AS: at 298.15 K, and these values will continue to be appropriate at some higher temperature T. Hence, AG: at the higher temperature, AGz(T), is given by AG:(T')
= M z ( 2 9 8 . 1 5 K)
-
TAS:(298.15
(17.1)
K)
As an illustration, consider the reaction 2Ag(s) +
; 0 2 w
(17.2)
= Ag20(s>
As Table 15.1 shows, at 298.15 K, AG: is -1 1.2 kJ mol-l: this means that Ag2O is thermodynamically stable with respect to silver and oxygen. Will AS:
for this reaction be positive or negative?
There is a decrease in the number of moles of gas, so AS: By referring to Equation 17.1, estimate whether AG: increase or decrease with increasing temperature.
is negative.
for Reaction 17.2 will
Consider the right-hand side of Equation 17.I . The first term, AH: (298.15 K), is a constant: it does not change with temperature. Because ASg(298.15 K) is negative, the second term, --TAS: (298.15 K), is positive. As T increases, the second term will therefore become more positive. We conclude therefore that AG: for Reaction 17.2 will become more positive as T increases. At some higher temperature, AG: will change from a negative to a positive quantity. At this temperature, Ag2O will cease to be thermodynamically stable with respect to silver and oxygen, and it will become thermodynamically unstable. If the reaction is fast enough, decomposition will then occur. The changeover temperature, the temperature at which AG: becomes zero, is called the thermodynamic decomposition temperature. Use Equation 17.1 to obtain a formula for this decomposition temperature. When AG:(T)
= 0,
Mg(298.15 K) = TAS:(298.15
T=
K)
AHg(298.15K) AS:(298.15K)
Table 17.1 Thermodynamic data for the silver-oxygen reaction at 298.15 K
Table 17.1 lists data for the silver-oxygen reaction at 298.15 K. Using these in Equation 17.3, we obtain -3 1.O kJ mol-' 3 1000 J mol-I T = = 466K -66.5 J K-' mol-I 66.5 J K-l mol-' Notice how AH:, which is quoted in tables in kJ mol-1, must be converted to J mol-1 to match the J in the unit of AS.:
116
(17.3)
This prediction is confirmed experimentally: when it is heated, silver oxide does indeed decompose at about 466 K. For Reaction 17.2, AGE increases with temperature because ME(298.15 K) is negative. This conclusion is perfectly general: when the temperature is increased, AGE for any reaction becomes more positive if AS: is negative, and more negative if AS: is positive. Now look again at Table 15.1. Will AG: for all these reactions become more positive as the temperature increases? Yes. For each reaction there is a decrease in the number of moles of gas: AS: will be negative in all cases. In Figure 17.5 we plot AGE for selected reactions from Table 15.1 against 7'.Such a plot is called an Ellingham diagram. As we predicted, in each case AGE becomes more positive as T increases. Ag20 appears right at the top of the Figure, and AGE does become zero close to our calculated temperature of 466 K.
Figure 17.5 An Ellingham diagram showing values of AGE for the formation of metal oxides from the metallic element and half a mole of 02(g) between 298 K and 2 000 K. Blue circles represent the points at which the metals boil, green circles the points at which they melt. The red squares on the vertical axes are for the reaction C(graphite) + :02(g) = CO(g)
Immediately below the plot for Ag20 is the one for HgO. In this case, AGE becomes zero at about 750 K. You can see that the decomposition temperature would be higher than this, were it not for a steepening of the slope above the blue circle at about 630 K. Blue circles in Figure 17.5 represent points at which the metals boil: at 630 K, the mercury in the equation (17.4) Hg + ; 0 2 @ = HgOW changes from a liquid to a gas. 117
Why does this increase the upward slope of the AGE plot? Above 630 K, there is a decrease of 1.5 mol of gas in the equation, so AS: becomes more negative. The rate of increase of AGE with T therefore becomes greater. At about 750 K, AGE is zero and HgO decomposes to oxygen gas and mercury vapour.
HgO and Ag2O undergo thermal decomposition at lowish temperatures because they are near the bottom of Table 15.1. The AG: values for their formation reactions are not very negative, so the increase in AG: with temperature (a consequence of negative AS: values) brings AGE to zero at a relatively low value of T. However, this is not the case for the other oxides in Figure 17.5. These decompose at inconveniently high temperatures, well above 1 000 K. To reduce these oxides to the metal, some method other than thermal decomposition must be used.
Because of the way our Ellingham diagram is plotted, values of AGE for the oxides at the top of Table 15.1 occur at the bottom of Figure 17.5. This means that if, at some temperature is the AGE plot for the oxide of metal A lies below that for the oxide of metal B, then metal A is thermodynamically capable of reducing the oxide of metal B at that temperature. Now the lines in Figure 17.5 tend to have similar slopes, except where the boiling of metals causes an abrupt change *. This means that the relative strengths of the metals as reducing agents change little with temperature: calcium, for example, is thermodynamically capable of reducing all the other oxides, both at 298 K and 2 000 K. We can attribute the similar slopes to similar negative values of AS: : all the formation reactions, when the metal is solid, involve a decrease of half a mole of oxygen gas. When the metal becomes gaseous, there is then a total decrease of 1.5 mol. Now consider the gas carbon monoxide, whose formation reaction is C(S> + I
1
3)2(g) = cow
(17.5)
If a plot for this oxide is inserted in Figure 17.5, will its slope resemble that of the other oxides? In the formation reaction, there is an increase of half a mole of gas, so A S : is positive and AGE decreases with temperature. The plot for carbon monoxide on the Ellingham diagram will therefore slope downwards from left to right.
You can obtain this plot by joining the two red squares at the extreme left and right of Figure 17.5. i
Join them with a pencilled line. If carbon forms CO(g), how many of the oxides is it thermodynamically capable of reducing at 298.15 K? Three. On the left-hand axis, only silver oxide, mercury oxide and copper oxide lie above carbon monoxide.
* The melting of metals also causes a change in slope, but as Figure 17.5 shows, the change is very slight. 118
How many of the oxides is carbon thermodynamically capable of reducing at 2 000 K? At 2 000 K, all of the oxide plots except those for CaO and A1203 lie above the carbon monoxide line. Hence all except these oxides can be reduced by carbon at this temperature.
When carbon is used to reduce oxides, the temperature is usually such that CO(g) is produced rather than CO2(g). As we have seen, carbon then becomes a more and more powerful reducing agent as the temperature is raised. The ability of carbon to reduce magnesium oxide, provided the reaction temperature is about 2 000 K, is especially striking. However, such temperatures require expensive and sophisticated furnace technology: consequently, carbon reduction is normally used at operating temperatures below about 1 750 K. At this temperature, carbon going to carbon monoxide will still reduce the oxides of lead, zinc and, most important of all, iron. The deliberate use of carbon to reduce oxygen compounds of copper and iron seems to have occurred first somewhere in eastern Europe or the Middle East. For copper, the date was about 4500BC; for iron, about 1200BC. With the possible exception of agriculture, a more important technological step along the road to modern civilization can hardly be imagined (Figure 17.6). It should now be clear to you that this technology is based on a unique property of carbon: it is a highly involatile element which forms volatile oxides. This combination yields the positive value (89.4 J K-' mol-*) of for the carbon monoxide formation reaction.
Figure 17.6 A reconstruction of primitive iron production.
We have now covered the reduction of the oxides of metals in our first and second classes. In the third class are the remaining metals, which include the alkali and alkaline earth metals, and aluminium. These have oxides that either require inconveniently high furnace temperatures for carbon reduction, or are difficult to prepare and handle. Electrolytic methods are widely used, often on fused halides. It should now be clear that, because of the broad similarities between different series, like those set out in Tables 12.1 and 15.1, and in the answer to Question 15.2, 119
Figure 17.7 An approximate classification of metals by extraction processes.
it is possible to say that, in a thermodynamic sense, some metals such as the alkali and alkaline earth metals are much more difficult to extract from their ores than others. This statement can be made without paying detailed attention to the precise nature of the ore being discussed. The classes 1-3 described at the beginning of Section 17 serve to arrange groups of important metals in order of increasing difficulty of extraction, and relate certain practical features of metal extraction to thermodynamic series such as that given in Table 15.1. In Figure 17.7, an attempt has been made to extend the classification to all the metals in the Periodic Table; it is only approximate. Broadly speaking, the generalizations made in Figure 17.7 are correct, but some metals require rather specific extraction methods that defy attempts at straightforward generalizations. The shading key to Figure 17.7 is: Red These metals occur towards the top of series such as those in Tables 12.1 and 15.1. They are usually extracted by the electrolysis of fused halides, or by heating oxides or halides with a strongly reducing metal such as calcium. For example, some alkali and alkaline earth metals are obtained by the electrolysis of fused chlorides. The lanthanide metals and actinide metals, such as uranium, are extracted by heating fluorides with calcium. All these metals usually dissolve quickly in dilute acids, forming positive ions and evolving hydrogen gas. Green Most of these metals are extracted by carbon reduction of oxides. They nearly all (note the exception of copper) dissolve slowly in dilute hydrochloric acid at room temperature, forming positive ions and evolving hydrogen gas. Yellow These metals occur towards the bottom of series such as those in Tables 12.1 and 15.1, They occur in the elemental form, or as ores from which the metal can usually be obtained by moderate heating in air. They do not dissolve in dilute hydrochloric acid.
I20
1
2
3
In Equation 11.4, AH: and AGg do not vary much with temperature, so, to a good approximation, we can assume that (17.1) AGz(7) = M g ( 2 9 8 . 1 5 K) - TASE(298.15 K) This equation enables us to predict how AG: will vary with temperature. An Ellingham diagram for metal oxides shows how AG: for the formation of oxides from half a mole of oxygen and the metallic element varies with T. The plots usually have positive slopes because AS: is negative. As carbon monoxide is a gas, AS: for this oxide is positive. The AG: plot has a negative slope, and so carbon is a very powerful reducing agent at high temperatures.
Use the thermodynamic data at 298.15 K in the Data Book to estimate the thermodynamic decomposition temperature for the breakdown of red HgO into mercury vapour and oxygen: HgOW = Hg(g) + ;om Compare your value with the one implied by Figure 17.5.
(17.6)
Use Figure 17.5 to find the approximate temperatures at which carbon being oxidized to CO(g) will reduce MnO and ZnO. Write an equation for the reaction in the zinc case, specifying the reactants and products at the reduction temperature.
Tick the appropriate column to show whether the following statements are true or false: Thermodynamically speaking, copper is harder to extract from its oxide than tin. Carbon going to carbon monoxide becomes a better reducing agent for metal oxides as the temperature is raised. The lanthanide and actinide elements are found free in nature. Calcium metal should be a good agent for reducing the compounds of other metals, whether those compounds are oxides, fluorides, chlorides, bromides or iodides. The alkali metals are obtained by heating their oxides or chlorides. Electrolysis is often used to extract the metals that form compounds with the most negative values of AGF.
121
In Tables 12.1 and 15.1, and in the chloride series in the answer to Question 15.2, metals were arranged in order of their thermodynamic tendency to combine with particular oxidizing agents. In Section 15, we pointed out that these series differed in detail, but also noted that there were marked overall similarities between the series, for example the high positions of the alkali metals, and the low positions of copper, silver and gold. What is it about the alkali metals that gives them a high position in each series? We shall now try to answer this, the last of our four introductory questions. Let us compare sodium and silver chlorides. At 298.15 K Na(s) + $12(g) = NaCl(s); A G : = -384kJmol-1 Ag(s)
+
iCl,(g) = AgCl(s); AGE = -1lOkJmol-1
(5.2) (18.1)
As in the other series, AGE is more negative for the sodium than for the silver reaction, and sodium lies above silver. Now the two reactions are analogous reactions in the sense defined in Section 16. The TASE values will therefore be similar, and because
AGE = AH:
-
TAS:
(11.4)
the difference in the AG* values for the two reactions will be almost entirely due to the difference in AHEmThe Data Book gives us AH: values for the two reactions: 1 Na(s) + ,C12(g) = NaCl(s); AH:= -411 kJmol-1 (5.2) Ag(s) + ;C12(g) = AgCl(s); AH:
= -127 kJmol-1
In what way do the AGE and AH: TAS: terms will be similar?
values bear out our assertion that the two
(18.1)
The difference in the AG: values (274 kJ mol-l) is similar to the difference in the AH: values (284 kJ mol-l). It follows that to explain why sodium lies above silver in the chloride series of Question 15.2, our main problem is to explain why AH: is more negative for the sodium reaction. We shall try to do this by relating the AH: values to the thermodynamic properties of sodium and silver atoms. You know that one mole of sodium metal and half a mole of chlorine gas react directly to give one mole of sodium chloride via Equation 5.2. However, we shall now imagine an alternative, indirect route at 298.15 K, which takes place in steps of our own choosing, and consider the enthalpy change for each step. These steps are as follows: 1 The metallic bonds in the sodium metal are broken and the metal is converted into gaseous sodium atoms: (18.2) Na(s) = Na(g) The standard enthalpy change for this step is positive and is called the standard enthalpy of atomization of sodium, AH& (Na, s). It is 107 kJ mol-l. 122
2
The bonds in our half a mole of C12 molecules are broken, converting the molecules into chlorine atoms: 1
yClz(g) = CKg)
(18.3)
The standard enthalpy change for this step is positive. When one mole of C12 molecules decompose in this way, the standard enthalpy change is C12(g) = 2Cl(g); AH: = 244kJmol-' (18.4) It is then called the bond dissociation energy of the chlorine molecule and given the symbol D(C1-Cl). For Reaction 18.3, the standard enthalpy change is halfthe bond 1 dissociation energy of the chlorine molecule, zD(Cl-Cl). This is 122 kJ mol-l. 3 Now an electron is removed from each gaseous sodium atom: (18.5) W g ) = Na+(g) + e-(g) The standard enthalpy change at 298.15 K for this step is positive ;it is referred to as the first ionization energy of sodium, Z1(Na), and its value is 496 kJ mol-l. 4 Now the electrons from the sodium atoms are added to the gaseous chlorine atoms to form gaseous chloride ions: CKg) + e-(g) = c1-(g)
(18.6)
Here you have to be careful. In Reaction 18.6, heat is evolved, and traditionally, the energies of it, and reactions like it, are defined in terms of an electron afJinity, in this case the electron affinity of chlorine, E(C1), which is the amount of heat evolved when Reaction 18.6 takes place. Here too therefore, the electron affinity of an element is defined as the heat evolved when a gaseous atom of the element accepts an electron at 298.15 K. Now we require the enthalpy change of Reaction 18.6, which, from Equation 6.1 is the heat absorbed. This is equal to minus the heat evolved, and as the electron affinity is the heat evolved, AH: = -E(Cl). The electron affinity of chlorine is 349 kJ mol-'. What is the value of AH: Reaction 18.6?
for
AH: = -E(CI) = - 3 4 9 ~ m o l - 1 5
The overall effect of steps 1 4 has been to convert sodium metal and chlorine gas into the gaseous ions, Na+(g) and Cl-(g). This can be seen by adding Equations 18.2, 18.3, 18.5 and 18.6: Na(s)
+ iC12(g) = Na+(g) + Cl-(g)
(18.7)
So the final step on the way to NaCl(s) is the one in which these gaseous ions come together to form solid crystalline sodium chloride: Na+(g) + Cl-(g) = NaCl(s) (18.8) The standard enthalpy change of this reaction at 298.15 K is called the lattice energy of sodium chloride and given the symbol L(NaC1, s). Equation 18.8 is a process of bond formation prompted by the attractive forces between oppositely charged ions, so it is one in which heat is evolved. Thus, L(NaC1, s) is negative, but how negative is it? We can find out by representing what we have done in a thermodynamic cycle. This particular one is known as the Born-Haber cycle (Figure 18.1).
123
Figure 18.1 Born-Haber cycle for a solid univalent metal chloride, MC1.
A solid chloride, MCl, is formed from the metal and chlorine either by the single equation at the base of the cycle with an arrow in the anticlockwise direction or by the clockwise sequence of steps 1-5 that have been described above. Where M is sodium or silver, all the terms in the cycle have been individually determined by experimental methods except for the lattice energy, L. The values are shown in Table 18.1.
Table 18.1 Terms in the Born-Haber cycles for sodium and silver chlorides
Write down an equation that relates AHy(MC1, s) to the other terms in the cycle. The cycle describes the formation of MC1 from the metal and chlorine, either by the equation at the base of the cycle with an arrow in the anticlockwise direction, or by the five-step clockwise route described in this Section. By Hess's law, the enthalpies of the two routes must be equal. So AHy(MC1, S) = AHZm(M,S)
+ Zl(M) + iD(Cl-Cl)
- E(C1) + L(MC1, S) (18.9)
Now use the data in Table 18.1 to calculate the lattice energy of sodium chloride. From Equation 18.9, the top five numbers in Table 18.1 must add up to the bottom number. Thus, for NaC1: (107 + 496 + 122 - 349) kJ mol-1 + L(NaC1, s) = -41 1 kJ mol-l L(NaC1, s) = -787 kJ mol-' If this calculation is repeated with the data for silver, one finds that L(AgC1, s) = -916 kJ mol-'. Pencil those two lattice energies into the blank spaces in Table 18.1.
124
You should now have a complete Table 18.1, which shows how AH? values for NaCl(s) and AgCl(s) can each be obtained by adding up five quantities. Of these five 1 quantities, two, D(C1-Cl) and -E(Cl), are identical for both NaCl and AgC1. This 2 leaves AHZm(M,s), Il(M) and L(MC1, s). Which of these three quantities tend to make for AgCl?
AH: more negative for NaCl than
The standard enthalpy of atomization and the first ionization energy of the metal: AH&(Na, s) is 178 kJ mol-' less than AHzm(Ag, s); and II(Na) is 235 kJ mol-' less than II(Ag). Equation 18.9 shows that both these terms tend to make AHF(NaC1, s) more negative than AH?(AgCl, s). The other term, the lattice energy, tends to produce the opposite effect, because it is 129 kJ mol-1 more negative for AgCl than for NaCl. Thus, if we parcel up the enthalpies of formation of the two chlorides in the way that we have done in this Section, we find that it is mainly the larger value of the sum ( AHzm(M, s) + 11) which causes silver chloride to lie below sodium chloride in a series based on the values of AH: for Reactions 5.2 and 18.1. As the two values of TASE are so similar, silver chloride also lies below sodium chloride in a series based on the values of AG:. Now suppose that we changed the reactant that combined with the metal M in Figure 18.1 from chlorine to, say, fluorine, and redrew the cycle. What thermodynamic quantities in the cycle would remain unchanged? The terms that are properties of the metal alone, AHZ (M, s) and I1(M). If we consider a change from NaCl to NaF, i D ( C I C 1 ) would change to 1 -D(F-F), E(C1) would change to E(F), and L(NaC1, s) would change to 2 L(NaF, s). However, AHZm(Na, s) and Z1(Na) would remain the same. It follows that, in a reconstituted Table 18.1 that compared NaF with AgF, the top two rows, which give AHZ(M, s) and Il(M) values, would remain the same. These values would now tend to make AHy(NaF, s) more negative than AHy(AgF, s). If we generalize this observation, we see that the (M, s) and I1(M)terms exert such an influence on the relative values of the standard enthalpies of formation, that there will be a strong tendency for sodium to be above silver whether the metals react with chlorine, fluorine, or any other standard reactant. Basically, then, the alkali and alkaline earth metals lie above noble metals such as copper, mercury, silver and gold in Tables 12.1 and 15.1, because they have lower atomization enthalpies and lower ionization energies. As we switch from one Table to another, variations in lattice energies produce important changes in detail, but the noble metals always lie near the bottom, and the alkali and alkaline earth metals are always near the top.
AHZ
In conclusion then, metals with relatively low enthalpies of atomization and relatively low ionization energies are those that are hardest to extract from their compounds. They are also the metals that most readily reduce other metallic compounds or ions.
125
1
2
The Born-Haber cycle enables us to calculate the lattice energies of halides from experimental data such as the standard enthalpies of atomization and ionization energies of metals, the dissociation energies and electron affinities of halogens and the enthalpies of formation of the halides. Analysis via the Born-Haber cycle suggests that metals with relatively low enthalpies of atomization and ionization energies are hardest to extract from their compounds.
Figure 18.2 shows the Born-Haber cycle for lithium fluoride. The quantities a-f represent enthalpy changes for reactions occurring in the direction of the arrows. From the quantities listed below, choose those that should be assigned to the six steps a-f in the cycle: the dissociation enthalpy of fluorine, D(F-F); 1 one-half of the dissociation enthalpy of fluorine, ZD(F-F); the electron affinity of fluorine, E(F); minus the electron affinity of fluorine, -E(F); - AH? (LiF, s); AHy(LiF, s); the first ionization energy of the lithium atom, Z,(Li); (viii) minus the lattice energy of lithium fluoride, -L(LiF, s); (ix) the lattice energy of lithium fluoride, L(LiF, s); (x) the standard enthalpy of atomization of lithium metal, AHZm(Li, s).
Figure 18.2 Born-Haber cycle for lithium fluoride, LiE
Use the following data to calculate the lattice energy of lithium fluoride: AH$m(Li, s) = 159 kJmol-I; I , (Li) = 520 kJ mol-I; D(F-F) = 158 kJ mol-I; E(F) = 328 kJmol-I; AHy(LiF, s) = -616 kJ mol-'.
126
The value of AH? (CaC12,s) is -796 kJ mol-l. For calcium, the standard enthalpy of atomization of the metal is 178 kJ mol-', and the first and second ionization energies are 590 and 1 145kJ mol-', respectively. Use these data, and data from Table 18.1, to calculate a value for the lattice energy of solid CaC12. Compare your value with that for NaCl in Section 18, and comment on the difference.
Look at Figure 17.7. Broadly speaking, would you expect those metals with the highest ionization energies to be to the left or the right of Figure 17.7?
127
We have now completed our introduction to thermodynamics and, at the same time, responded to the problem of the activity series of metals. Metals such as sodium and calcium are powerful reducing agents, and hard to extract from their ores. They have this status because of their relatively low enthalpies of atomization and ionization energies. The chief purpose of the rest of this Book is to begin a study of the chemistry of the typical elements. So between Section 23 and the end of the Book, you will find an account of the chemistry of the elements of Group I and Group I1 he alkali and alkaline earth metals. But before we can embark on the descriptive chemistry, there are three thermodynamic quantities that you need to know more about. You have already been introduced to two of them- the lattice energy and the ionization energies of elements, but we need to add some more details. The third quantity is called the redox potential, and it provides an ‘alternative currency’ for the compiling of Table 12.I . Sections 20-22 are devoted to these three thermodynamic quantities.
128
In Section 18 and Question 18.3, we used the Born-Haber cycle to calculate the lattice energies of sodium and calcium chlorides, L(NaC1, s) and L(CaC12, s). They are equal to the standard enthalpy changes of the following reactions: Na+(g) + C1-(g) = NaCl(s); AH: Ca2+(g) + 2Cl-(g) = CaC12(s); AH:
= -787 kJmol-1
= -2 255 kJ mol-1
(20.1) (20.2)
In these reactions, gaseous ions condense to form a solid. No one has yet succeeded in determining the energies of such changes by direct measurements on reactions such as 20.1 and 20.2. The only way of obtaining experimental values of lattice energies is through the Born-Haber cycle. In this Section we shall seek reasons why these important quantities have the values that they do. Sodium and calcium chlorides are usually treated as ionic compounds: they consist of charged ions. To pull oppositely charged ions such as Na+ and C1- apart against the attractive force of the ions requires work (energy). So the separation of ions is an endothermic process and AH is positive. How does this explain the negative lattice energies of NaCl and CaC12 in Equations 20.1 and 20.2? When the gaseous, oppositely charged ions are brought together to form a solid, energy is released and the process is exothermic. These arguments are applications of an ionic model (Figure 20. I), a model which assumes that certain compounds are composed of ions. They are also only qualitative. But in the next Section we shall express them quantitatively.
(b) Figure 20.1 The ionic model represents a crystal (a) of halite (sodium chloride) as an assembly of oppositely charged ions, like that used in (b) to represent the structure’s unit cell.
129
A quantitative application of the ionic model is discussed in The Third Dimension2. The underlying assumption was that ions could be represented by hard charged spheres. This assumption allows a set of ionic radii to be developed, and then used to predict likely crystal structures from radius ratios. We shall use the same notion of hard charged spheres here. Suppose we have two such ions (e.g. Na+ and C1-) with equal and opposite single charges +e and -e; the centres of the ions are separated by a distance r (Figure 20.2). Because of the opposite charges, there is a Coulomb force of attraction between the ions. It then turns out that the electrostatic energy, Ec, arising from this Coulomb attraction is given by: eL (20.3) /yc = - 4mor Here e is the electronic charge, 1.6 x 10-19 C, and Q is a known physical constant called the permittivity of a vacuum. Thus, on the top of the fraction in Equation 20.3 we have the product of the equal and opposite charges, +e and -e; on the bottom, we have a constant ( 4 7 ~ multiplied ~) by the internuclear distance, r.
Figure 20.2 Two hard oppositely charged spheres, whose centres are separated by a distance Y.
Now imagine that the two charged spheres are separated by an enormous distanceso large that it may be regarded as infinite. What then is the value of Ec? Zero; the fraction in Equation 20.3 consists of a quantity, -e2, divided by 4 7 ~ ~ 0 r , where r is infinitely large. It will therefore be minute -effectively zero. Now suppose the two ions, initially separated by this huge distance, move progessively closer together. The separation Y gets smaller, so the quantity (e2/47qr) will increase from zero. But in Equation 20.3, this quantity has a negative sign in front of it, so Ec, from being zero, will gradually become more negative. The steady decrease in Ec occurs naturally because the equal and opposite charges are pulled together by an attractive force, so they tend to approach each other.
130
When does this natural approach of the two ions stop? We are modelling our ions as hard spheres. It will stop when they make contact with each other (Figure 20.3).
Figure 20.3 Two oppositely charged ions, represented1 as hard spheres, in contact with each other. The internuclear distance Y is now the sum of their radii.
You can think of the spheres as balls made of hard rubber. When they make contact, the steady decrease in potential energy caused by the decrease in Ec stops. Any further decrease in the internuclear distance r must be brought about by pressing the spheres together against the repulsive force of the hard rubber. This requires us to do work on them, so there is now a steep increase in the energy of interaction due to a new term, the repulsive energy ER. The total energy for the two ions is therefore E = Ec 4- ER (20.4)
The combined effect of these two terms is shown in Figure 20.4. To the right, the Coulombic energy, Ec, is dominant and as the internuclear distance, r, diminishes, E becomes more negative. To the left, after the ions have come into contact, E R is dominant and E increases. In between, there is an equilibrium internuclear distance ro, where the total energy of interaction, E, is a minimum, Emin. This is the most stable situation, and the internuclear distance observed at this energy minimum corresponds to that found in crystals.
Figure 20.4 How the total interaction energy, E, of two oppositely charged ions varies with internuclear distance. 131
In Equation 20.4, we have the two terms Ec and ER on the right. We already have a quantitative expression for Ec in Equation 20.3. To go further, we need one for ER also. Max Born (Figure 20.5) suggested that one possible answer has the form B (20.5) ER =F where B is a constant, and n (which is known as the Born exponent) is a number lying between 5 and 12. Thus, ER is proportional to Urn, where n is much greater than one. This is why ERincreases so steeply at short internuclear distances at the left of the minimum in Figure 20.4. We can now substitute Equations 20.3 and 20.5 into Equation 20.4: e2 E=-+-B (20.6) 4 m 0 r rn It is this equation which gives us the curve shown in Figure 20.4. But we are interested in the energy of the minimum, Emin.This corresponds to the lattice energy, in that it gives us the energy of the ion pair at the equilibrium internuclear distance ro, relative to the zero energy they possess as fully separated ions in the gas phase. There are well-established mathematical techniques for finding Emin,and at the same time eliminating the constant B. The result is Emin . =-- e2 (1 - 1 (20.7) 4m0r0 n
->
So far, so good: we have an expression (Equation 20.7) for the energy of interaction of a pair of singly charged ions. But an ionic crystal is more complicated in three important respects: 1 It contains more than two ions! Indeed, we have calculated lattice energies in kJ mol-l. The Avogadro constant, N A , tells us the number of formula units in one mole. It has the value 6.022 x 1023mol-l. So when calculating the lattice energy of sodium chloride, we are thinking of crystals containing 6.022 x 1023 Na+ ions and 6.022 x 1023 C1- ions. 132
2
3
Our ion pair contained singly charged ions, but in general our ionic crystal may contain ions with other integral charges such as +3e or -2e. Different ionic crystals have different unit cells, and therefore different arrangements of ions in space. We must expect the expression for the lattice energy to change from one type of structure to another (Figure 20.6).
Allowing for these three complications is mathematically quite difficult, but the result is surprisingly simple. To obtain the lattice energy of an ionic crystal, all that is necessary is to multiply Equation 20.7 by a quantity NAMZ+Z-. Hence
L = - NAMZ+Z- e2 (1 - -11 (20.8) 4m0r0 n This is called the Born-Lande' equation. We have already explained the meaning of the 0 , e and n. N A is the Avogadro constant, equal to 6.022 x 1023mol-l; symbols 4 7 ~ ~ro, this is the allowance for complication 1 above. Z+ is the number of positive charges +e on the positive ions in the crystal, and Z- is the number of negative charges -e on the negative ions; together they allow for complication 2 . Finally, M is a number that is characteristic of the particular crystal structure under consideration. It is called the Madelung constant and allows for complication 3. Some values of M are shown for various crystal structures in Table 20.1. We shall have more to say about them later. Table 20.1 Some crystal data relevant to the calculation of lattice energies
Figure 20.6 A crystal of fluorite, CaF2,may
contain some ions. The ions have different charges (+2e and -e), and a specific arrangement in space. All this must be allowed for when transforming Equation 20.7 into an expression for the lattice energy of the crystal.
Equation 20.8 can be simplified by substituting numbers or physical constants for NA, e2 and 4xq. The result is LkJmol-'
=-
1.389 x 105MZ+Zr0
1 (1 - -)Pm n
(20.9)
Note the units kJmol-l and pm that appear on the left and right, respectively, of the equation. They tell us that if we insert values of ro in picometres (pm), we obtain lattice energies, L, in kJ mol-l.
133
We shall now use Equation 20.9 to calculate the lattice energies of the compounds LiF, NaCl and CaC12. We shall then compare our results with the experimental values obtained by using the Born-Haber cycle. These experimental values were calculated in Section 18, and in Questions 18.2 and 18.3. They appear in the last column of Table 20.2, along with the quantities that we shall need to substitute into Equation 20.9. Table 20.2 Calculating the lattice energies of NaC1, NaF and CaC12 from the Born-Land6 equation
As the Table shows, NaCl and LiF both have the sodium chloride structure, and CaC12 has the rutile structure. We can therefore enter values of the Madelung constant M from Table 20.2. Remember that M multiplies up the interaction of the simple ion pair to allow for the fact that we are dealing with the interactions of a collection of ions arranged in a crystal. It is larger for the rutile than for the sodium chloride structure. This is mainly because a mole of Ti02 or CaC12 crystals contains 50% more ions than a mole of NaCl crystals. Next in Table 20.2 come the numbers of positive or negative charges on the ions. Notice that these numbers are all positive. There are, for example, two positive charges on each Ca2+ ion in CaC12 and one negative charge on each C1- ion. Next come the values of n, the Born exponent. They can be obtained from experiments that study the effect of pressure on the volumes of crystals. However, in most cases we shall be dealing with ions with noble gas configurations. Approximate values of n can then be obtained from the numbers given by Pauling, which appear in Table 20.3. A constant is given for each noble gas configuration. To obtain a value of n for an ionic crystal, one takes the average for the noble gas configurations of the ions in the formula unit. Thus, in NaCl, the Na+ ion has the configuration of neon (n = 7), and the C1- ion that of argon (n = 9), so the average is n = 8. This is the value of n for NaCl which appears in Table 20.2. What are the values of n for LiF and CaC12?
6 and 9, respectively; in LiF, the Li+ and F- ions have the helium (n = 5 ) and neon (n = 7) configurations so the average is 6. In CaC12, there are three ions to average over, but all of them have the argon configuration so n = 9. Enter these values of 6 and 9 in Table 20.3.
134
Table 20.3 Constants used to calculate the Born exponent, n, for crystals containing ions with noble gas configurations
Finally, Table 20.2 contains the values of ro, the equilibrium internuclear distance that can be obtained by X-ray diffraction. We now have all the data that we need to calculate L from Equation 20.9. For NaCl,
1 1.389 x lo5 x 1.748 x 1 X 1 (1 - -)kJmol-l 28 1 8 = -756 kJ mol-1
L(NaC1, s ) =
Now do Question 20.1.
Use Table 20.2, and the data that you have entered in it, to calculate the lattice energies of LiF and CaC12. Check your results against those given in this Table. The last two columns of Table 20.2 suggest that Equation 20.9 works pretty well. The values are less negative than the experimental values obtained from the BornHaber cycle, but only by about 4% or less. Our simple combination of just two terms, Ec and ER,has ‘captured’ some 96% of the lattice energy. There are also possible reasons why the calculated values are not negative enough. We have, for example, not allowed for the London (or van der Waals) forces between the ions. Inclusion of these would give more negative values of L and improve the agreement. It looks, therefore, as if our hard-sphere ionic model might well provide us with an understanding of the energies of ionic compounds and their reactions. Let us take it a little further.
Look again at Equation 20.9:
LkJmol-I = -
I 1.389 x 1 0 5 ~ z + z (1 - -)Pm r0 n
(20.9)
To apply it quantitatively to a particular compound, we need the Madelung constant M , which varies from one crystal structure to another. But there are occasions in chemistry when we do not need to be as quantitative as this. It is then useful to have a simpler equation that merely gives us an idea of how lattice energies change from one ionic compound to another, regardless of any change in crystal structure. Table 20.1 sets us on the road to this simpler equation. Column 3 contains the Madelung constants of various crystal structures, and column 4, the numbers of ions, v, in each formula unit. Thus, in A1203, there are two Al3+ ions and three 02ions, so v = 5. The values of M vary substantially, but in column 5 , they have been divided by V . What does division by v do to the variation in M? It largely disappears. The values of Mlv span the small range 0.80-0.88. This was first noticed by the Russian chemist, A. F. Kapustinskii. What it suggests is that it will not be too bad an approximation to replace M in Equation 20.9 by a number in the range 0.80-0.88, multiplied by v. As Equation 20.9 tends, if anything, to give values that are not quite negative enough, it makes sense to choose a number 135
in the upper part of the range. We therefore choose the NaCl value of 0.874and replace M by 0.874~. Furthermore, you know of a set of ionic radii that can be used to predict internuclear distances in ionic compounds. So we can replace ro by (r+ + r-), where r+ and r- are the radii of the positive and negative ions. The result of these two changes is: LkJmol-' = -
1.214x 105vz+ z1 (1 - ->pm r+ + r-
n
(20.10)
One further step is needed. The term (1 - l/n)in Equation 20.10consists of two parts. The second part, lln, is only a small fraction of the first, reflecting the fact that the repulsive energy ER is only a small fraction of the Coulombic energy, Ec. We can therefore approximate this l/n term without much reduction in accuracy. As n varies between 5 and 12 (Table 20.3),Kapustinskii used the value 9. The sum total of these changes gives us the Kapustinskii equation: L=-
wvz+z-
(20.11)
r+ + r-
where W = 1.079 x 105 kJ mol-1 pm. For NaC1, LiF and CaC12, using the ionic radii given in the Data Book (on the CD-ROM), this equation gives lattice energies of -763,-1 033 and -2 304 kJ mol-I, respectively. If you look at Table 20.2,you will see that its performance compares very favourably with the more rigorous Born-Land6 equation, although when a wider range of compounds is studied, it does not do so well. It will, however, prove very useful in qualitative arguments. For example, the equation shows clearly that L(LiF, s) is more negative than L(NaC1, s) because the ions in LiF are smaller than those in NaCl. Their centres are therefore closer together in the crystal.
When we draw the Born-Haber cycle for NaCl (Figure 20.7),it includes the step I
-c12(g) 2 = Cl(g); AH:
= 122 kJmol-1
(18.3)
As C12(g) is the standard state of elemental chlorine, and gaseous chlorine atoms are formed in Equation 18.3,we could write the standard enthalpy change of this step as AHY(C1, g). If you look in the Data Book, you will see that AHy(C1, g) is indeed 122 kJ mol-1. However, the break up of the molecules in the gas phase can be described in terms of bond dissociation energies. As the C12 molecules are gaseous,
Figure 20.7 The Born-Haber cycle for NaCl. 136
1
AHy(C1, g) is also equal to 5 D(C1-Cl), half the bond dissociation energy of C12. It is usual to write the step in this more specific form, as in Figure 20.7. 1
When we move to NaBr and NaI, this equality between AHy(X, g) and D(X-X) no longer exists. At room temperature, bromine is a liquid and iodine a solid: the standard state of these halogens are Br2(1)and 12(s),respectively. The Born-Haber cycle for NaI takes the form shown in Figure 20.8. Again, there is a step corresponding to Equation 18.3: 1
212(s) +I(&;
AH: = 107 kJ mol-1
(20.12)
Again also, AH: is equal to AHy(1, g), as you can verify in the Data Book. 1 However, this time it is not equal to TD(I-I). This is because the diatomic molecule on the left is in the solid rather than the gaseous phase.
Figure 20.8 The Born-Haber cycle for NaI.
The precise relationship between
AH:
(I, g) and
1
D(1-I) is revealed by Figure 20.9.
Figure 20.9 A thermodynamic cycle for the conversion of solid iodine to gaseous atoms.
Along the top, Reaction 20.12 takes place in two steps. Firstly, the half-mole of solid iodine is vapourized into gaseous I2 molecules; the standard enthalpy change is AHY(I2, g). From the Data Book this is 31 kJ mol-1. Then the gaseous 12 molecules are taken apart into gaseous atoms. It is this step whose standard enthalpy 1 change is TD(I-I), with a value of 76kJmol-l. The total energy change from left to right is (3 1 + 76) or 107 kJ mol-1 , and it is this which is equal to AH (I, g).
7
When doing Born-Haber cycle calculations in general, it is best to use a cycle of the type shown in Figure 20.8 with AH (X, g) values from the Data Book. But you need to be aware of the possible use of bond dissociation energies because notably for fluorides and chlorides -the data will sometimes be given to you in this form.
7
137
1
A hard-sphere ionic model leads to an expression for the lattice energy of an ionic crystal. This is the Born-Land6 equation:
n Here, M is a constant called the Madelung constant, which is characteristic of the particular crystal structure. Z+ and 2- are the numbers of charges on the positive and negative ions, n is a number known as the Born exponent, and ro is the internuclear distance between the ions. The Born-Land6 equation gives L values that agree quite closely with the experimental ones obtained from the Born-Haber cycle. The Born-Land6 equation can be further simplified into the Kapustinskii equation. This ignores differences in crystal structure, but nevertheless gives a useful idea of how lattice energies vary with the sizes and charges of the ions in ionic compounds. It takes the form YO
2
3
r+ + rHere, W is a constant equal to 1.079 x 105 kJ mol-1 pm, v is the number of ions in the formula unit, and r+ and r- are the radii of the positive and negative ions.
Draw a Born-Haber cycle for KBr, and use it to write down an equation for the lattice energy of this compound in terms of ionization energy, electron affinity, etc. Substitute data from the Data Book into the equation, and thus calculate L(KBr, s).
KBr has the sodium chloride structure, with an internuclear distance of 330 pm. Use the Born-Land6 equation to calculate its lattice energy. How well does your value agree with the experimental value that you obtained in Question 20.2?
Estimate the Born exponent, n, for CaF2, which has the fluorite structure, with an internuclear distance of 237 pm. Calculate L(CaF2, s) from the Born-Land6 equation. How well does your value compare with the experimental figure of -2 635 kJ mol-1 obtained from the Born-Haber cycle?
The values for the ionic radii of Ca2+and F- given in the Data Book are 114 pm and 119 pm, respectively. Use the Kapustinskii equation to calculate L(CaF2, s). Is the agreement with the experimental value better or worse than that obtained with the Born-Land6 equation?
138
In Table 12.1, we graded metals in terms of their strengths as reducing agents in aqueous solution. Suppose a metal M forms an aqueous ion Mrz+(aq). What then is the reaction whose AG: 1
;M(s)
value appears in Table 12.1?
+ H+(aq) = $Mn+(aq)+ iH,(g).
Thus, for copper, the ion is Cu2+(aq)and n = 2, so we have: 1
-Cu(s) 2
1 + H+(aq) = ZCu2+(aq) + iH2(g); AG:
= 32.7kJmol-1
(21.1)
In Table 12.1, the metals with the most negative values of AGE were placed at the top, and those with the most positive values at the bottom. The strengths of the metals as reducing agents then decrease as you descend the Table. When assembling Table 12.1, we emphasized the method by which AG: is obtained by combining the separate AH: and AS: values as (AH: - TAS:). However, in certain cases, AG: for reactions such as Reaction 2 1.1 can be obtained directly by electrical measurements. Consider Figure 21.1. The left-hand beaker contains a platinum metal electrode in contact with hydrogen gas and H+(aq). The right-hand beaker contains a copper metal electrode in contact with Cu2+(aq). The two beakers are connected by a glass tube containing a conducting solution of KCl(aq); this is known as a salt bridge.
Figure 21.1 An electrochemical cell in which Reaction 21.2 occurs spontaneously.
This arrangement creates a voltage between the electrodes. If, as in Figure 21.1, the terminals are joined by a wire and small resistor, electrons flow in the wire from left to right. The electrode from which the electrons flow is marked negative; the electrode towards which they flow is marked positive. This electron flow is provided by reactions that occur at each electrode. These are shown in Table 21.1.
139
Table 21.1 Electrode reactions in an electrochemical cell
(21.2)
The arrangement shown in Figure 21.1 is called an electrochemical cell. It is shown in discharge mode; that is to say, it is releasing energy or doing electrical work by pushing an electric current through the resistor. The energy needed to do this comes from the overall chemical reaction that takes place in the cell. You exploit the same principle every day when you use electrical batteries (Figure 21.2), although these develop much higher voltages than the cell shown in Figure 2 1.1. This is discussed further in the Batteries Case Study. The Gibbs energy change of this cell reaction can be determined from the voltage developed by the cell. This voltage can be found by replacing the resistor with a counter-voltage that can be adjusted to stop the natural flow of electrons in Figure 21.1, This ‘stopping voltage’ is equal to the maximum possible voltage that the cell can deliver. It is known as the electromotive force (e.m.f.), E , of the cell. If we arrange things so that reactants and products in the total reaction are all in their standard states (Section 9.2), it is the standard e.m.f., Ee of the cell. In the case of Figure 21.1, the stopping voltage when this is so is found to be 0.339 V. By convention, the e.m.f., in this case p,is obtained by giving this stopping voltage the sign of the electrode on the right. So what is the E? value for the cell of Figure 21.1 ?
The electrode on the right is the copper electrode, which has a positive sign; ,?P= +0.339V. Now we must find the cell reaction to which this p value refers. The rule is that it must be written so that oxidation occurs at the left-hand electrode, and reduction at the right. In the case of Figure 2 1.1, that very reaction appears in Table 2 1.1: at the left-hand electrode, hydrogen is oxidized to hydrogen ions, and copper ions are reduced at the right. Thus, E? = +0.339 V refers to: (21.3) Cu2+(aq) + H2(g) = Cu(s) + 2H+(aq); I!? = 0.339 V
Now that the cell reaction has been properly written down, we can find its value of AG: from an equation that we state without proof AG: = - n F p (21.4) Here, n is the number of electrons that are gained by the oxidized state at the left (Cu2+) as it changes to the reduced state on the right (Cu). In this case, therefore, n = 2. F is a constant called the Faraday constant, whose value is 96.485 kJ mol-I V-’ . Therefore (21.5) AG: = -2 x 96.485 x 0.339 kJ mol-1 = -65.4 kJ mol-1 Thus, for the cell reaction (Equation 21.3), we find Cu2+(aq) + H&) = Cu(s) + 2H+(aq); AGE = -65.4 kJmol-I
140
(21.6)
Figure 21.2 Some examples of batteries containing different chemicals. In
discharge mode, these chemicals react and supply energy.
Is this value consistent with that taken from Table 12.1 and shown in Equation 2 1.1? Yes; if Equation 21.6 is reversed and halved, it becomes 1
-Cu(s) 2
+ H+(aq) =
1
ICu2+(aq)
+
1
,H*(g); AGg = 32.7 kJ mol-1
(21.1)
So far we have only described the measurement of the standard e.m.f. of the cell in which-the right-hand beaker of Figure 21.1 contains copper metal and Cu2+(aq); see Equation 21.3. But clearly, we could obtain the standard e.m.f. of other cell reactions of the same type by replacing Cu(s) and Cu2+(aq)by a different metal and its aqueous ion; for example (21.7) Ag+(aq) + ;H2(g) = Ag(s) + H+(aq); I!? = 0.80 V
Zn2+(aq) + H2(g) = Zn(s)
+ 2H+(aq); I!? = -0.76V
(21.8)
To abbreviate cell reactions 21.3, 21.7 and 21.8, we now replace each combination 1 of ?H2(g) on the left with H+(aq) on the right by the symbol e. This gives: Cu2+(aq) + 2e = Cu(s);
= 0.34V
(21.9)
Ag+(aq) + e = A&); @ = 0.80V
(2 1.10)
Zn2+(aq) + 2e = Zn(s); ?I
= -0.76V
(21.
Table 21.2 Standard redox potentials at 298.15 K
These values are called standard redox potentials. Table 21.2 contains a fuller list arranged in ascending order of p. There are four important points to note about this Table: The symbol ‘e’ in each equation is not strictly an electron: if it were, it would 1 have a superscript minus as in e-. The symbol e is shorthand for ,H2(g) on the side on which it appears, and H+(aq) on the other side. However, if you think of it as an electron, each reaction in the table becomes balanced with respect to charge. In all of the reactions, the e symbol appears along with an oxidized state or oxidizing agent (e.g. Mg2+) on the left; a reduced state or reducing agent (e.g. Mg) appears on the right. Thus a concise way of expressing equations such as 2 1.11 is to write @(Zn2+ I Zn) = -0.76 V, with the oxidized state first inside the bracket, and the reduced state second, with a vertical bar separating the two states. AG: for these equations can then be obtained from Equation 21.4. The more powerful the reducing agent in a redox system, the more negative is the I!? value. Roughly speaking, redox systems with I!? values more negative than about -0.1 V contain powerful reducing agents and weak oxidizing agents; redox systems with ?!I values more positive than about 1.1 V contain powerful oxidizing agents and weak reducing agents. Redox potentials can be used to predict whether or not a particular redox reaction is thermodynamically possible. Because the system with the more negative value contains the stronger reducing agent, we can make the following generalization:
141
Electrochemical cells are a very convenient way of determining AG: values in aqueous solution, but their application is limited. Often cells cannot be persuaded to develop their full e.m.f., and some electrode systems, like those of the alkali metals, are not viable because one of their components reacts with water. The value can then be found by the non-electrical methods described earlier in this Book: AG,8 for the redox reaction is determined from AH: and AS:, and then is calculated from Equation 2 1.4. Question 2 1.1 provides an example of this.
Aluminium is one of the metals whose standard redox potential cannot be determined by an electrochemical measurement. In the Data Book, AGF(A13+,aq) = -485 kJ mol-l. This is the value of AG: for the following reaction: Al(s)
+ 3H+(aq) = A13+(aq)+ ;Hz(g); AG:
= -485 kJmol-l
(21.12)
Use this to obtain a value of AGE for the process written as A13+(aq)+ 3e = Al(s) (2 1.13) and then use your value to calculate p ( A l 3 + I Al). Does your result suggest that aluminium is a strong reducing agent?
(a) What are the two strongest and two weakest oxidizing agents in Table 21.2? (b) What are the two strongest and two weakest reducing agents in Table 21.2?
Use the information in Table 21.2 to predict which of the following reactions are thermodynamically favourable: (i) reduction of Zn2+(aq)by magnesium metal; (ii) reduction of Sn2+(aq)by zinc metal; (iii) reduction of Ca2+(aq)by zinc metal; (iv) oxidation of Fe2+(aq)by C12(g); (v) oxidation of Br(aq) by C12(g). Where a reaction occurs, write an equation for it.
142
In Section 19, we promised that Sections 20-22 would deal with three thermodynamic quantities. The lattice energy (Section 20) drew attention to ions in the solid state; the redox potential (Section 21) focused on ions in aqueous solution. Our third quantity, the ionization energy, is concerned with gaseous ions and their formation from gaseous atoms. In Section 18, the ionization energy of a chemical species A was defined as the molar enthalpy change at 298.15 K, for the reaction: (22.1) A(g) = A+@ + e-(g) In other words, it is the molar enthalpy change for the removal of an electron from A. Figure 22.2 shows how the first ionization energies of the atoms in Figure 22.1 (and hydrogen) vary with increasing atomic number. What is the general nature of the change that occurs across each Period from Group I to Group VIII? There is an overall increase in ionization energy. Compare, for example, the values for lithium and neon. Let’s try to relate this change to the change in electronic configuration. At helium, the electronic configuration of the atom is ls2, and at lithium it is ls22sl. We can therefore write the latter as [He]2sl, where [He] denotes the helium configuration. At neon, at the right-hand end of Period 2, the configuration is [He]2s22p6.Thus,
Figure 22.1 The mini-Periodic Table, showing the typical elements.
Figure 22.2 The first ionization energies, Z1,of the elements in Figure 22.1, plotted against atomic number. Note the breaks on the horizontal axes where the transition series and the lanthanides have been omitted. 143
in moving across the Period from lithium to neon, one proton is added to the nucleus at each step, and one electron is added to the shell with principal quantum number n = 2. Let’s consider these two effects in turn. Concentrate first on just one of the electrons in the shell with n = 2, and imagine that this is the one that is removed on ionization. What overall effect should the successive addition of protons have on the ionization energy from lithium to neon? It should produce an overall increase: the build up of positive charge on the nucleus means that the electron is subjected to an increasingly strong attractive force; this suggests that progressively more energy will be needed to remove the electron from the atom. Now we turn to the second effect. What overall effect should the successive addition of electrons to the shell with n = 2 have on the ionization energy from lithium to neon? It should produce an overall decrease: the build up of accompanying electrons in the shell means that the original electron is subjected to an increasingly strong repulsive force; this tends to make ionization progressively easier. Thus, the two effects compete with each other: one tends to produce an overall increase across the Period, and the other an overall decrease. Which effect turns out to be dominant? The increase in the nuclear charge: the experimental results in Figure 22.2 show that this effect overwhelms the influence of electron repulsion to produce an overall increase in ionization energy across the Period. This dominance of the increasing nuclear charge is perhaps not surprising since it acts from the centre, whereas the increasing electron repulsion arises from electrons that are dispersed in shells around the nucleus. Having accounted for the overall increase, let’s now turn to the deviations that occur from it. Figure 22.3 shows the variation from lithium to neon in an enlarged form. The upward trend is broken by downward displacements between beryllium and boron, and between nitrogen and oxygen. How would you relate the break between beryllium and boron to the change in electronic configuration? It marks the point where the electrons stop entering an s sub-shell, and begin filling up a p sub-shell. In explaining the overall increase in ionization energy between lithium and neon, we assumed that, as the electron that is lost is always in the shell with n = 2, it is in the same state at each element. But this is not so: for the elements lithium and beryllium, the electron being lost is in a 2s sub-shell; for the elements boron to neon, it is in a 2p sub-shell. It is therefore not surprising that the plot of the ionization energies for the second set of elements is displaced with respect to that for
144
Figure 22.3 The first ionization energies of the elements of Period 2, which runs from lithium to neon.
the first. As the displacement is downwards, the implication is that, other things being equal, there is something about s electrons that makes them harder to remove than p electrons. This point is taken further in EEements of the p Block3. The second downward break, between nitrogen and oxygen, can be related to Hund’s rule. Sketch the box diagrams for the electronic ground states of the elements boron to neon. How do oxygen, fluorine and neon differ from boron, carbon and nitrogen? The 2p sub-shells in the box diagrams are shown overleaf (Figure 22.4). Unlike those of boron, carbon and nitrogen, the 2p sub-shells of oxygen, fluorine and neon contain paired electrons. On ionization, it is one of these paired electrons that is lost. Thus, the ionization of an oxygen, fluorine or neon atom leads to a reduction in the number of paired electrons, whereas the ionization of a boron, carbon or nitrogen atom does not. Now Hund’s rule tells us that the ground state of an atom is the one that maximizes the number of parallel spins in an unfilled sub-shell. This marks the fact that pairs of electrons with opposed spins in a sub-shell repel one another more strongly than do pairs with parallel spins. Thus, an ionization that eliminates paired spins is favoured relative to one that does not. Consequently, in Figure 22.2, the plot for 0, F and Ne is displaced downwards with respect to that for B, C and N.
If you examine Figure 22.2, you will see that the breaks between Groups I1 and 111, and between Groups V and VI, appear in some later Periods as well, although in a less marked form. Our argument shows that those breaks separate atoms that contain np electrons from those that contain just ns electrons, and atoms that contain p electrons with opposed spins from those that do not.
You have just seen how the increase in nuclear charge leads to an overall increase in ionization energy between lithium and neon. If the 2p sub-shell could contain more than six electrons, then one would expect sodium, the element that follows neon, to have a higher ionization energy than any element in the lithium-neon Period. However, as Figure 22.2 shows, there is a steep drop after neon; indeed, it is so steep that the ionization energy of sodium is lower than that of lithium, or indeed any other element in Period 2. The electronic configurations of the lithium and sodium atoms are [He]2s1 and [Ne]3s1, respectively. Why, despite its much higher nuclear charge, does the sodium atom have the lower ionization energy? The sodium electron lost on ionization is in the shell with n = 3. The electrons in this shell tend to be much further from the nucleus than those in the shell with n = 2. This, presumably, is the reason why the outer electron in sodium is held less strongly than that in lithium, even though the sodium atom has the greater nuclear charge. This decrease in ionization energy between lithium, and sodium, the element below it in the same Group, illustrates a general trend. The ionization energies of the typical elements tend to decrease down a Group as the outer electrons occupy shells of higher principal quantum number which are further from the nucleus. The tendency is discernible in Figure 22.2: compare the values for the halogens and noble gases. It is even clearer in Figure 22.5, which also reveals a small number of exceptions. Although the sizes of these exceptions look small, their consequences are quite important; they are taken up in EEements of the p BZuck3.
Figure 22.5 The change in first ionization energy down each of the eight Groups in Figure 22.1. 146
Figure 22.4 Box diagrams for the electronic structures of the atoms boron to neon.
1
2
3
The first ionization energies of the atoms of the typical elements show an overaZZ increase across each Period, and an overaZZ decrease down each Group. The overall increase across each Period is broken at points where p electrons first join s electrons in the outer electronic configuration, and where p electrons in the box diagram can no longer all have parallel spins. The overall increase in ionization energy across each Period shows the importance of the effect of increasing nuclear charge. This binds the outer electrons more and more tightly across each row. The overall decrease in ionization energy down each Group, shows the importance of the increase in distance of the outer electrons from the nucleus as the successive electron shells build up. This increasing distance weakens the attractive force between the nucleus and the outer electrons.
By referring to Figure 22.1 only, state which atom of each of the following pairs of elements has the higher first ionization energy: (i) aluminium and sulfur; (ii) nitrogen and antimony; (iii) germanium and chlorine.
147
We now begin at the descriptive chemistry of the typical elements. The rest of this Book is taken up with Groups I and I1 (the other Groups are dealt with in Elements of the p BZock3). Here, we start with Group I, the alkali metals. Figure 23.1 shows their position in the Periodic Table. Most of the compounds that they form are usually regarded as ionic, and treated by an ionic model. In Sections 18-22 we discussed Born-Haber cycles, lattice energies and ionization energies. These concepts are highly relevant to the ionic model and will therefore prove useful in understanding alkali metal chemistry. Of the five common alkali elements, lithium, rubidium and caesium are all much less common than sodium or potassium, which are, respectively, the seventh and eighth most abundant elements in the Earth’s crust. In many parts of the world, there are vast deposits of sodium chloride, which have been formed by the evaporation of ancient seas (Figure 23.2). At present, world industry consumes about 190 million tonnes of sodium chloride each year, but this is merely a drop in the dried-out ocean of the Earth’s total reserves. The UK’s source in Cheshire contains more than 100 000 million tonnes, and much larger deposits occur elsewhere.
Figure 23.2 The Wieliczca salt mines near Cracow in Poland include an underground chapel with chandeliers made from crystalline salt.
148
Figure 23.1 The alkali metals in the Periodic Table.
The alkali metals occur near the top of Tables 12.1, 15.1, 15.2 and 21.2. This shows that, thermodynamically speaking, they have a strong tendency to be oxidized. What does this suggest about the difficulty of extracting such metals from their compounds? It will be considerable; if the metals are readily oxidized, the resulting ions or compounds will be difficult to reduce. In Section 17.3, you saw that metals of this type are obtained, either by reducing their compounds with another highly oxidizable metal such as calcium, or by electrolysis. Therefore it is not surprising that lithium and sodium are obtained by electrolysis of their molten chlorides, and that rubidium and caesium are made by the reduction of their chlorides by calcium metal. The five alkali metals have the body-centred cubic structure (Figure 23.3). Some physical properties of the metals, their atoms and their ions are shown in Table 23.1.
Figure 23.3 Metallic lithium, sodium, potassium, rubidium and caesium have a bodycentred cubic structure.
Table 23.1 Some properties of the alkali metal atoms and ions
The alkali metals have unusually low densities, melting temperatures and boiling temperatures. For example, lithium, sodium and potassium are less dense than water: during their vigorous reactions with water, they float (Figure 23.4). In Section 1.1 we discussed the simple electron gas model of metallic bonding; this is pictured again in Figure 23.5. The atoms of the metallic elements form positive ions and free electrons, and the latter are dispersed throughout the structure. The dispersed electrons act as a sort of glue: their attraction for the oppositely charged positive ions holds the ions together in the metallic structure. Use this model to suggest why the alkali metals have low melting temperatures and low densities.
Figure 23.4 The low density of sodium is revealed by the fact that it floats during its reaction with water. 149
The alkali metals form cations with a single positive charge. Thus, the attraction between the electron gas and the ion cores is weaker than in other metals, and the positive ions are not pulled together so strongly. The densities and melting temperatures are therefore lower.
To reinforce this point, Table 23.2 compares the relevant properties of sodium with those of magnesium and aluminium, the two adjacent metals in the Periodic Table (Figure 23.6). Table 23.2 Some properties of metallic sodium, magnesium and aluminium, and of their atoms
Figure 23.5 The electron gas model of metallic bonding.
Using the electron gas model, what are the expected charges on the ion cores in metallic magnesium and aluminium? They are +2 and +3, respectively. As with Na+, there is a tendency to form ions with a noble gas configuration (neon); these ions are Mg2+ and Al3+. With these charges, the increase in density and melting temperature from sodium through magnesium to aluminium is what we would expect. Another mark of the unusually weak bonding in the alkali metals is the low value of AH:m(M, s), the standard enthalpy change of the reaction: M(s) = M(g) (23.1) Table 23.2 also shows that the first ionization energy of sodium is lower than that of magnesium or aluminium; considering the family as a whole, the Group I elements have lower first ionization energies than any other Group in the Periodic Table. As you saw in Section 22, this is because the nuclear charge that acts on a particular electron shell is lowest at the beginning of the Period, where the Group I elements occur. Now, when the alkali metals form free M+ ions, the metal is first broken down into atoms (Equation 23. l), and then the gaseous atom is ionized: (23.2) M(g) = M+(g) + e-(& The relatively low enthalpy changes for Equations 23.1 and 23.2 therefore show that the alkali metals are more readily converted into gaseous M+ ions than any other group of metals. In Section 18 we showed how the low values of AH% (M, s) and I1(M) also accounted for the readiness of the metals to be oxidized to compounds or sohtions containing M+ ions; for example
+ :C12(g) = MCl(s) M(s) + H20(1) = M+(aq) + HO-(aq) +
(23.3)
M(s)
150
iH&)
(23.4)
Figure 23.6
The relative positions of the alkali metals, magnesium and aluminium in the Periodic Table of Figure 22.1.
In this case, readiness to be oxidized was understood in a thermodynamic sense: the low values of AHzm(M, s) and Z1(M) tend to make Reactions 23.3 and 23.4 thermodynamically very favourable. However, this by itself cannot account for the explosive nature of such reactions; a second, complementary condition is required. What condition is this? A kinetic one: the reactions must not only be thermodynamically very favourable, but they must also befast. The reactions of the alkali metals tend to be fast because of the low values of AH$m(M, s). The rate of the reaction is limited by the speed with which the metal structure can be broken down; a metal with a low value of AHzm(M, s) is easily taken apart. If this is so, use Table 23.1 to decide which alkali metal should react most slowly. The values of AH$m(M, s) decrease down the Group; lithium has the largest value and should therefore react most slowly. As you know, when the alkali metals are dropped into water, lithium undergoes a relatively mild reaction. Although the alkali metals readily form compounds or solutions containing M+ ions, the ionization never goes further to give substances containing M2+ions. Tables 23.1 and 23.2 show why: the second ionization energies of the alkali metals are enormous -much higher than those of magnesium or aluminium. Why is 12 for an alkali metal such as sodium so large? The second ionization energy, Z2, is the ionization energy of the ion, Na+. This ion has the noble gas structure of the neon atom. On ionization, an electron must be removed from an inner shell (principal quantum number n = 2) of the sodium atom whose electrons are much closer to the nuclear charge. The very low first ionization energies and very high second ionization energies of the alkali metals are a quantitative indication of the stability of the noble gas electronic configuration, which plays such an important role in the elementary bonding theories that you should be familiar with. The low first ionization energies suggest that the noble gas configuration is relatively easily formed from the metals; the high second ionization energies show that it is broken into only with great difficulty.
23.1. I The alkali metals in liquid ammonia Table 23.1 contains the standard redox potentials of the alkali metals in water, P ( M + I M). These are all large and negative. They put the alkali metals at the top of Table 21.2, and show that they are very powerful reducing agents in aqueous solution. Here we use a very remarkable example to show that this reducing power is evident in other solvents also. Ammonia gas, NH3, like water, is the hydride of a non-metal. It can be liquefied by cooling it below its boiling temperature of -33 "C. Like water, the liquid is colourless. If an alkali metal is added to liquid ammonia in small amounts, the metal dissolves and a blue solution is formed (Figure 23.7). 151
The alkali metal forms colourless positive ions, M+(amm), where ‘amm’ signifies that they are dissolved in ammonia; the blue colour is due to electrons, which like the positive ions, are dispersed throughout the ammonia solvent: M(s) = M+(amm) + e-(amm) (23.5) What has happened to the alkali metal in this reaction? It has undergone oxidation; it has lost an electron. How does Reaction 23.5 differ from a typical redox reaction? The alkali metal has been oxidized, but nothing seems to have been reduced. Oxidation and reduction are usually complementary processes. What has happened here is that the electrons that have been lost have not been passed to an oxidizing agent. They have been ‘parked’ in the solvent, where they are awaiting something that they can reduce. Not surprisingly, therefore, solutions of the alkali metals in liquid ammonia find many uses as strong reducing agents in both organic and inorganic chemistry. Reactions of type 23.5 are rare; apart from the alkali metals, the only metals that undergo them are the Group I1 elements, calcium, strontium and barium, and two of the lanthanide elements, europium and ytterbium. These, like the alkali metals, are all metals with a high tendency to be oxidized.
152
Figure 23.7 The blue colour of ammoniated electrons deve:loping as sodium dissolves in liquid ammonia.
1
2
3
4
5
The alkali metals have low melting temperatures and boiling temperatures, and are very easily oxidized. They react readily with water giving alkaline solutions of hydroxides and hydrogen gas. The vigour of this reaction increases down the Group. The metals also dissolve in liquid ammonia, giving blue solutions containing the ammoniated electron. When the alkali metals are oxidized in these and other reactions, compounds or solutions containing M+ ions are formed. In alkali metal atoms, the nuclear charge is lower than in other elements in the same Period. The first ionization energies, in which a noble gas configuration is formed, are therefore relatively low. By contrast, the second ionization energies are unusually large because an electron must be removed from an inner shell of the atom where electrons are closer to the nucleus than the outer s1 electron. The combination of low first, and high second ionization energies is primarily responsible for the readiness of the alkali metals to form substances containing M+ ions, but not ions of any higher charge. According to the simple electron-gas model of metallic bonding, the alkali metals contain only one freed electron per atom. Thus, the interaction between the ion core and electron gas is relatively weak. This accounts for the softness, and for the low densities, melting temperatures and atomization energies of the alkali metals. Because the value of AHZm(M,s) is low, the structure of an alkali metal is easily broken down and the metals tend to react relatively quickly.
The values of AH: and AGE for the reactions of lithium and caesium with water are very similar. However, caesium reacts explosively and lithium steadily; explain the difference in terms of the properties of the atoms and their ions.
153
Most alkali metal compounds are quite soluble in water. This is especially the case with compounds of sodium, the most common of the five elements. When the compounds dissolve, their positively charged ions, such as Na+(aq), are formed, and these are very hard to reduce, and impossible to oxidize. Thus one of the most important uses of sodium is as an inert cation (the ‘counter ion’) that can carry negatively charged ions into solution. For example, few carbonates or hydroxides, apart from those of the alkali metals, dissolve easily in water, yet solutions containing the ions OH-(aq) or C03’-(aq) have many practical uses both in industry and elsewhere. Thus, sodium carbonate, Na2C03,and sodium hydroxide, NaOH, are produced by the chemical industry in very large amounts (Figure 24.1). Their value as sources of hydroxide and carbonate ions is enhanced by the resistance of the accompanying sodium ions to oxidation or reduction: there is no chance that chemicals that are meant to react with hydroxide or carbonate will react with the sodium ions instead.
Figure 24.1 The undistinguished appearance of sodium carbonate (left) and sodium hydroxide (right) gives no hint of their huge industrial importance.
The importance of sodium carbonate and sodium hydroxide is apparent from the fact that about 60% of the sodium chloride consumed in chemical manufacturing worldwide is used in a sector known as the chlor-alkali industry. The major products of this sector are chlorine, sodium hydroxide and sodium carbonate. To give you a flavour of the industry we shall simply describe the most modern method of making two of its three chief products: sodium hydroxide and chlorine. 154
Sodium hydroxide and chlorine are obtained commercially by the electrolysis of concentrated sodium chloride solutions. It sounds simple, but there is an important problem that must be solved first. To appreciate this, look first at the simplest possible arrangement, in which two electrodes are immersed in a concentrated solution of sodium chloride, and connected to a battery (Figure 24.2). Write down an equation for what happens at the positive electrode. The equation is: Cl-(aq) = ;cl2(g)
+ e-
(24.1)
Negative chloride ions in the solution migrate to the positive electrode, surrender their electrons, and are then evolved as chlorine gas.
Figure 24.2 The simplest possible way of electrolysing a concentrated solution of sodium chloride.
Now for the negative electrode. An over-hasty judgement might be that here the positive sodium ions might be discharged as sodium metal: (24.2) Na+(aq) + e- = Na(s) This is incorrect, but by starting in this way we can arrive at the correct reaction. Suppose that sodium metal were formed in this way. Write an equation for what would happen to it. The equation is: Na(s)
+ H20(1) = Na+(aq) + OH-(aq) +
kH,(g)
(24.3)
Because water is present, sodium metal will react with it in the usual fashion. Thus, the overall process at the negative electrode would be the sum of Equations 24.2 and 24.3. Add them up, and write down their sum. Both the sodium ions, and sodium metal cancel out, leaving H20(1)
+ e-
= OH-(aq)
+
iH2(g)
(24.4)
To summarize, the initial reactions during the electrolysis of sodium chloride are as follows: positive electrode: CI-(aq) = kclz(g) + e(24.1) negative electrode: H20(1) + e- = OH-(aq) + iH;?(g) (24.4) Because the sodium ions in the solution remain unchanged, the build-up of hydroxide ions at the negative electrode amounts to a build-up of aqueous sodium hydroxide solution. Thus, the initial products are chlorine gas at the positive electrode and sodium hydroxide solution at the negative electrode. In principle, therefore, both our desired products can be obtained from one electrolytic cell. However, there is a danger that they will indulge in mutual destruction because hydroxide ions and chlorine gas react together to produce the chloride ion, C1-(aq), and the hypochlorite ion, C10-(aq): (24.5) Cl2(g) + 20H-(aq) = C1-(aq) + ClO-(aq) + H20(1) 155
It is therefore essential that the products produced at the positive and negative electrodes should be kept separate. The chemical industry has devised some ingenious ways of doing this.
24.1. I The membrane cell The separation of the chlorine and sodium hydroxide can be achieved by a membrane. The membrane allows the passage of the positive ion, Na+(aq), but not of negative ions such as C1-(aq) and HO-(aq). Figure 24.3 shows such a membrane cell. The positive electrode (left) is usually metallic titanium, often coated with ruthenium dioxide, RuO,; the negative electrode (right) is woven steel wire. Chlorine is formed in the left-hand compartment. If chloride ions could pass through the membrane from left to right, they would become unavailable for chlorine production and adulterate the sodium hydroxide solution that is formed around the negative electrode. The U-shaped arrow in the left compartment therefore marks these two valuable effects of the membrane.
Figure 24.3 Schematic diagram of the manufacture of sodium hydroxide and chlorine, using a membrane cell. Ideally, the membrane allows passage of positive ions only. Hydrox-ide ions cannot get into the positive compartment of the cell to destroy chlorine, and chloride ions cannot get into the negative compartment to contaminate the sodium hydroxide solution.
What valuable effect is marked by the U-shaped arrow in the right-hand compartment? The membrane stops OH-(aq) from entering the positive compartment and destroying chlorine via Equation 24.5. What sort of material is the membrane made of? It is an organic polymer whose basic skeleton has been fluorinated so that it resembles polytetrafluoroethene, (-CF2)n, which is also known as PTFE or teflon. However, in this instance the structure of the skeleton is modified at some points by bridging the (-CF2-) groups with oxygen atoms, for example (-CF2-O-CF2-). Groups of atoms are then attached to this basic skeleton at intervals in place of a fluorine atom. In what is known as the acid form of the polymer, these groups are -COOH and -S020H, 156
and when the membrane makes contact with an aqueous solution, hydrogen ions tend to be lost; for example, where the acid form contains a carboxyl group, the following equilibrium will be present 0 0
I
I
II = F-C-C-0-(mem)
II
F-C-C-0-H(mem)
I
I
+ H+(aq)
(24.6)
If the aqueous solution contains a high concentration of another positive ion such as Na+(aq), the negatively charged sites so created can be occupied by this ion:
I
0 It
F-C-C-0-(mem)
I
+
Na+(aq) =
0 I II F-C-C-0-Na
I
(mem)
(24.7)
Suppose such occupation can occur throughout the structure. Then, if the concentration of sodium is higher on one side of the membrane than the other, sodium ions can pass through the membrane from one site to another to reduce the concentration difference (Figure 24.4).
Figure 24.4 The membrane in a chlor-alkali membrane cell consists of a polymeric backbone based on polytetrafluoroethene. At intervals on this backbone there are -COOand -S020- sites, which can bind positive ions like Na+ and H+. Aqueous positive ions can therefore be transferred through the membrane from site to site, but aqueous negative ions cannot. For this reason, the membrane is known as a cation-exchange membrane.
Why can’t negative ions such as C1-(aq) and OH-(aq) enjoy the same experience? The membrane can offer only negatively charged sites; there are no positive sites where negative ions might be bound. Notice also that, because there is no bulk flow of water through the membrane, the water needed for the reaction at the negative electrode (Equation 24.4) cannot come from the brine added to the positive electrode compartment. So water is added separately to the negative electrode compartment. The brine concentration in the positive electrode compartment is maintained by combining a recycling of the depleted sodium chloride with a boost to its concentration (Figure 24.3). Solid sodium hydoxide is obtained by evaporation of the solution leaving the negative compartment.
The uses of sodium hydroxide in industry are shown in Figure 24.5. As we have seen, the most striking property of alkali metal hydroxides is their solubility in water, and the resulting ability to deliver high concentrations of OH-(aq). Hydroxide ions destroy acidity, in the shape of H+(aq), through the neutralization reaction: H+(aq) + OH-(aq) = H20(1) (24.8) This explains the sector labelled ‘neutralization’ in Figure 24.5.
Figure 24.5 Principal worldwide uses of sodium hydroxide. The use labelled ‘alumina’ refers to the preparation of pure aluminium oxide from bauxite during aluminium extraction (Section 3.3).
World industry produces some 30 million tonnes of sodium hydroxide each year, along with a similar quantity of sodium carbonate, Na2C03,the other sodium product of the chlor-alkali industry. A typical use-pattern of sodium carbonate is shown in Figure 24.6. As with sodium hydroxide, it is the alkalinity of the substance that explains its use in the manufacture of paper, soaps and detergents. We discuss its role in glass-making in Section 28.3.
Figure 24.6 The principal worldwide uses of sodium carbonate
158
1
2
The most important industrial compounds of the alkali metals are sodium hydroxide and sodium carbonate. Together with chlorine gas, these two substances are the major products of the chlor-alkali industry. Sodium hydroxide is produced at the negative electrode, and chlorine at the positive electrode, when brine is electrolysed. The two products must be kept separate because chlorine and hydroxide ions react together. In membrane cells, there is a cation-exchange membrane, with sites that permit the passage of only positive ions. Brine is then cycled through the positive electrode compartment, and the negative compartment must be topped up with water.
159
Alkali metal halides can be made by the direct reaction of the alkali metals and halogens. They are all white crystalline solids melting at high temperatures (450-1 000 "C) to give electrically conducting melts, and they all dissolve in water, yielding colourless solutions that conduct electricity; for example (25.1) NaI(s) = Na+(aq) + I-(aq) All the alkali metal halides have formulae of the type MX, where M is an alkali metal atom, and X is a halogen atom. However, because of these properties, so characteristic of ionic compounds, they are taken to consist of M+ and X- ions. Apart from CsC1, CsBr and CsI, all the alkali metal halides have the sodium chloride structure, whose unit cell is shown in Figure 25.1. Each sodium ion is octahedrally surrounded by six chloride ions. The compounds CsCl, CsBr and CsI have the caesium chloride structure (Figure 25.2), in which there are eight halides around each caesium. The alkali metal halides are substances of great theoretical importance. This is because the ionic model of chemical bonding is developed by constant reference to, and assessment against, the properties of the alkali metal halides. For example, they can be used to generate a set of ionic radii that help us to understand why caesium chloride has the structure that it does. The larger cation in the caesium halides
Figure 25.1 The unit cell of sodium chloride.
160
Figure 25.2 The unit cell of caesium chloride.
(see Table 23.1) provides the extra room needed to pack eight rather than six chloride, bromide or iodide ions around the metallic element. Likewise, in Section 20, the alkali metal halides were very useful in testing our attempts to use an ionic model for the prediction of lattice energies.
When the alkali metals are heated in plenty of oxygen, they combine with it to form compounds. We might expect such compounds to contain M+ and 0 2 - ions with the formula M20. In the case of lithium, this is quite correct. The oxide Li20 is formed, and it has the antifluorite structure, which has the unit cell shown in Figure 25.3. We may think of it as an assembly of Li+ and 02-ions. Each oxide ion is surrounded by eight lithium ions at the corners of a cube; each lithium ion is tetrahedrally coordinated to four oxide ions.
Figure 25.3 The unit cell of lithium oxide, Li20.
However, the compound formed by sodium is not Na20; it has the empirical formula NaO, where there is one oxygen for every sodium.
If this were a normal, ionic oxide, what would be the charge on the cation? A normal ionic oxide is taken to contain 02-ions; the cation would therefore be Na2+.
161
Figure 25.4 The structure of sodium peroxide, Na202: the oxygens are grouped in pairs with an internuclear distance of 149 pm, so the compound is assumed to contain peroxide ions with the formula 022-.
Now, we said in Section 23.1 that compounds containing Na2+ions are precluded by the very high second ionization energy of sodium. But this inconsistency is only apparent because the compound of empirical formula NaO is not a normal oxide; this is made clear by the crystal structure, which has the unit cell shown in Figure 25.4. In normal oxides, individual oxygen atoms are bound only to the atoms of other elements, as in Figure 25.3. In NaO, the oxygens are bound in pairs, and the distance between them is 149 pm, not much more than the distance 121 pm found in the O2 molecules of oxygen gas. The compound is therefore written Na202,and is called sodium peroxide: it contains Na+ cations and peroxide anions, 022-. With potassium, rubidium and caesium, the situation is different again. Whereas lithium and sodium in excess oxygen yield Li20 and Na202,potassium, rubidium and caesium form bright orange solids with the formulae KO2, Rb02 and Cs02. Again, the crystal structures show that these contain pairs of oxygens, separated in this case by only 128 pm. The compounds are called superoxides and are taken to be assemblies of M+ and the superoxide ion, 02-. The normal oxides, peroxides and superoxides of the alkali metals all react with water to give alkaline solutions of the hydroxides; for example K20(s) + H20(1) = 2K+(aq) + 20H-(aq) (25.2)
+ 2H20(1) = 2Na+(aq) + 20H-(aq) + H202(aq) 2K02(s) + 2H20(1) = 2K+(aq) + 20H-(aq) + H202(aq) + 02(g)
Na202(s)
(25.3) (25.4)
Notice that peroxides and superoxides give an aqueous solution of hydrogen peroxide, H202. This decomposes gradually to water and oxygen: H202(aq) = H20(1) + :02(g)
(25.5)
How many moles of oxygen gas, 02(g), are ultimately produced by the reaction of two moles of K02(s) with excess water? 1.5 mol; one mole is produced quickly by Equation 25.4, and another 0.5 mol by the slow subsequent decomposition of the mole of H202 also produced in the reaction. This 1.5 mol of 02(g) can also be released by direct reaction of the solid superoxide with carbon dioxide: 2K02(s> -k co2(g) = K2C03(S) + ;02(g) 162
(25.6)
Figure 25.5 In the Salyut 6 spacecraft, potassium superoxide, KO2, consumed exhaled C 0 2 and replaced it with oxygen.
What circumstances can you envisage in which this reaction, and therefore K02, would be very useful? A solid that can convert exhaled C02(g) into breathable oxygen could extend the lifetime of a life-support system in a spacecraft or submarine. Russian spacecraft (Figure 25.5) have, in fact, used superoxides for just this purpose.
25.2.1 The relative stability of peroxides and oxides Let's look more closely into why it is that when we burn lithium and sodium in air, lithium forms the normal oxide, Li20, but sodium yields the peroxide, Na202. Firstly, we should point out that lithium peroxide, Li202, does exist; however, it decomposes to the normal oxide on heating to about 200 "C: ~i202(s) = L ~ ~ O (+S >;o2(g)
(25.7)
Sodium peroxide, on the other hand, decomposes in this way only when heated to about 600 "C. Use Equation 25.7 to suggest why the products of lithium and sodium combustion are Liz0 and Na202, respectively. The temperatures generated by combustion lie below the decomposition temperature of Na202, but above that of Li202. Thus, the difference in the combustion products arises because the decomposition temperature of Na202 substantially exceeds that of Li202. We can now look at this problem thermodynamically. In Section 17.1, you were told that the thermodynamic decomposition temperature of a substance is given, to a good approximation, by T=
AH: (298.15 K) ASg(298.15 K)
(17.3)
163
where, in this case, AH: (298.15 K) and AS: reaction of the type M202W
(298.15 K) refer to a decomposition
= M20W + 12 0 2 k )
(25.8)
What can you say about the values of ASEfor Reaction 25.8 in the two cases (M = Li and M = Na)? The two reactions are analogous in the sense defined in Section 16; their AS: values will be very similar. What factor in Equation 17.3 leads to the decomposition temperature of Na202 being greater than that of Li202? Because the AS: values are similar, the AH: value must be larger for the sodium compound; this is so, as the data in Table 25.1 show. Table 25.1 Values of AH: and
m2 for Reaction 25.8 for M = Li and M = Na
Thus, if we can explain this difference in AH: values, we shall have explained why the decomposition temperature is higher for Na202than for Li202,and why the combustion products are Li20 for lithium and Na202for sodium. To obtain such an explanation, we use the thermodynamic cycle in Figure 25.6. The decomposition reaction is written along the top; this is the direct route from the peroxide to normal oxide plus oxygen; like all reactions in such cycles, the enthalpy change, AH:, refers to the reaction in the direction of the arrow.
Figure 25.6 Thermodynamic cycle for the thermal decomposition of alkali metal peroxides.
Now let us take an indirect route from peroxide to normal oxide by going round the cycle in an anticlockwise direction. There are three steps:
Step I The peroxide, M202, breaks down into its gaseous ions. Notice that this step is the reverse of the reaction in which the solid is formed from its gaseous ions and for which the standard enthalpy change is the lattice energy L(M202,s). Thus, the standard enthalpy change for this breakdown reaction is -L(M202, s).
164
Step 2 The gaseous ions, M+(g), appear on both sides of this second step so they can be ignored. The reaction then becomes one in which the gaseous peroxide ion breaks down into a gaseous oxide ion and oxygen gas: 022-(g) = 02-(g) + +(g)
(25.9)
The standard enthalpy change is simply labelled C in Figure 25.6.
Step 3 In this final step, the solid oxide M20(s) is formed from its gaseous ions; the standard enthalpy change is the lattice energy L(M20, s). Application of Hess's law shows that the enthalpy changes by the direct and indirect routes must be equal:
AH:
= -L(M202,
S)
+ L(M20, S) + C
(25.10)
Now we use the Kapustinskii equation, Equation 20.11 to obtain values for the lattice energies: (20.11) What are the values of v, Z+ and 2- for M202 and M20? In both cases, there are three ions in the formula unit, the cation charge is +1 and the anion charge is -2. Thus, v = 3,Z+ = 1 and 2- = 2, and 8
AHm =
6W 6w +c r(M+) + r(022-) r(Mf) + r(02-)
(25.11)
Now suppose we start with a decomposition reaction in which M is lithium, and we change to one in which M is sodium. Will AH: increase or decrease? Clearly, that will depend on how the three terms on the right-hand side of Equation 25.11 change. How will the third term, C, be affected by the change from lithium to sodium? It will not be affected at all. This is because the metals are not involved in Equation 25.9, for which C is the enthalpy change. Thus, the changes in AH: when sodium is substituted for lithium will depend only on the changes in the first two terms on the right-hand side of Equation 25.11 -that is, on the difference between 6W/[r(M+) + ~ ( 0 ~and ~ 6W/[r(M+) -)] + r(02-)]. Will 6W/[r(M+) + r(022-)]increase or decrease when M changes from Li to Na? It will decrease; Table 23.1 shows that the ionic radius of Na+, r(Na+), is greater than that of Li+, r(Li+)- 116pm against 90 pm. This increases the denominator of the fraction where the ionic radii appear. Likewise, the quantity that is subtracted from this, 6W/[r(M+) + r(02-)]will also decrease. Thus, whether AH: increases or decreases when M changes from Li to Na will depend on whether GW/[r(M+) + r(OZ2-)]or GW/[r(M+) + r ( 0 2 - ) ]decreases more. In deciding which, it is crucial to note that, because 022contains two oxygens and 02-contains only one, 022will take up a lot more space in its compounds than does 02-;so r(022-)will be considerably greater than r(02-). When M changes from Li to Na, which quantity decreases more, 6W/[r(M+) + r(022-)]or 6W/[r(M+) + r(02-)]?
165
6W/[r(M+) + r ( 0 2 - ) ] ;because r(02-) is smaller than 1 - ( 0 ~the ~ increase -), in r(M+)has the greater impact on this term.
If you find this difficult to follow, look at Figure 25.7, which takes you through an extreme, hypothetical case with real numbers. This also shows that because the , has a minus sign in front of it larger decrease occurs in 6W/[r(M+)+ r ( 0 2 - ) ]which in Equation 25.11, AH: becomes more positive when the cation size is increased.
Figure 25.7 A demonstration of the effect that dictates the change in AH: for the decomposition of an alkali metal peroxide when the size of the cation is increased. The expression for AH: includes a difference between two fractions with identical numerators, 6W. The denominators of both fractions are the sum of two terms, which we shall call (A + B ) for the first fraction and (A + C) for the second. A is common to both terms and B >> C. Illustrating with an extreme case, if B = 1 000 and C = 1, then when A is increased from 1 to 2, the decrease in the first fraction is negligible, but that in the second is much larger and results in the total expression becoming more positive.
The principles developed in this Section can be applied in many other situations. What we have shown is that the thermodynamic stabilities and decomposition temperatures of alkali metal peroxides increase with cation size. The argument depended on the fact that the anion in the decomposition product, 02-,was smaller Because of this, the value of -L for than the anion in the decomposing solid, 022-. the normal oxide product was greater than that for the decomposing peroxide, and it was also more substantially lowered by an increase in r(M+).
166
When heated quite gently in hydrogen gas, all of the alkali metals form colourless crystalline hydrides, MH. All five compounds have the sodium chloride structure shown in Figure 25.8. When lithium hydride, LiH, is heated, it melts at 680 O C , and if the melt is electrolysed, liquid lithium is formed at one electrode, and hydrogen gas at the other: negative electrode: Li+(melt) + e- = Li(1) (25.12)
positive electrode:
H-(melt) =
H,(g)
+ e-
(25.13)
Figure 25.8 The alkali metal hydrides all have the sodium chloride structure: in LiH, each lithium ion is octahedrally surrounded by six hydride ions, and each hydride ion is surrounded by six lithium ions.
This behaviour, together with the crystal structure, is characteristic of ionic compounds, and the alkali metal hydrides are usually formulated M+H-. When the alkali metal hydrides are dropped into water, they react to produce hydrogen gas; for example (25.14) LiH(s) + H20(1) = Li+(aq) + OH-(aq) + H,(g) During the Second World War, lithium hydride was sometimes used to inflate air-sea rescue equipment: Reaction 25.14 occurs on contact with seawater. All the alkali metal hydrides decompose into hydrogen and the metal when strongly heated: (25.15) M H = M + kH2 Lithium hydride, however, is by far the most stable with respect to this reaction. It melts at 680 OC, and decomposition begins at around 900-1 000 "C; by contrast, sodium and potassium hydride decompose at about 400 "C. This decrease in stability down the Group appears in an even more marked form when the metals are heated 167
in nitrogen gas. Only lithium reacts, forming a brick-red nitride, Li,N; the nitrides of other alkali metals are much less stable with respect to the metal and nitrogen. We shall not undertake a detailed analysis of the thermodynamics of these phenomena in the way that we did for oxygen compounds; instead we shall just summarize the essential conclusions. According to the Born-Haber cycle, when we compare salts of formula type M+X-, the factor that tends to stabilize the salt with respect to decomposition into its constituent elements is its lattice energy: other things being equal, the more negative the lattice energy (the larger -L), the more stable the salt will be. Now, according to Equation 20,11, the value of -L for an alkali metal hydride is 2W/[r(M+) + r(H-)I. The hydride ion, H- is one of the smallest negative ions, as one would expect of an anion containing only two electrons. Tables of ionic radii, including those in the Data Book, show that its ionic radius ( 1 26 pm) is similar to that of 02-. What does this suggest about the changes in -L for alkali metal hydrides when the cation radius, r(M+),increases? The value of 2W/[r(M+) + r(H-)] will fall because of the increase in r(M+), and because r(H-) is small for an anion, it will fall away particularly steeply. It is this steep fall in the value of -L(MH, s) that leads to the decrease in the stability of the alkali metal hydrides down Group I. Indeed, one can draw the general conclusion that the stability of alkali metal compounds containing small anions, with respect to their constituent elements, will tend to decrease down the Group.
The alkali metal halides are of formula type MX, and dissolve in water to yield M+(aq) and X-(aq) ions. They are colourless crystalline solids, which melt above 450 "C to give conducting melts. CsCl, CsBr and CsI have the eightcoordinate CsCl structure; the rest have the NaCl structure. When the alkali metals bum freely in air, lithium forms a normal oxide, Li20. ions, and potassHowever, sodium yields a peroxide, Na202,containing 022ium, rubidium and caesium form orange superoxides, M02, containing 02-ions. In water, all three types of oxygen compound form aqueous solutions of hydroxides; the peroxides also yield H202,and the superoxides form H202 and oxygen gas. The peroxides and superoxides combine with C02, liberating oxygen. Sodium peroxide decomposes to the oxide and oxygen at a higher temperature than Li202.This can be thermodynamically understood by considering -L(M202) and -L(M20), which are measures of the stabilities of the two types of oxygen is greater than compound with respect to their gaseous ions. Because 1-(0*~-) r(02-), -L(M202) is less than -L(M20), and decreases by a smaller amount when r(M+) increases. The preceding result can be generalized: if a solid ionic compound undergoes decomposition by breakdown of the anion into a solid compound with a larger -L value, the stability of the compound tends to increase with cation size. The alkali metals react with hydrogen to form colourless crystalline hydrides, M+H-, with the NaCl structure; these react with water to give aqueous hydrox168
ides and hydrogen gas. The stability of the hydrides to decomposition into the constituent elements decreases down the Group: KH and NaH decompose at around 400 OC, but LiH melts at about 700 "C without decomposition, to give a conducting melt that yields lithium and hydrogen at the electrodes during electrolysis.
7
When the metals are heated in nitrogen, only lithium forms a nitride. A major contributor to the decrease in the stabilities of the nitrides and hydrides down the Group is the small size of the anions, H- and N3-. This makes the decrease in -L(MH) and -L(M3N) with increasing cation size especially large.
Pure sodium peroxide is colourless, but the commercial product obtained by burning sodium in a free supply of air is pale yellow. When it is added to icecold water, it dissolves, and a small volume of oxygen gas is immediately evolved. This volume is increased some five- to tenfold if the solution is boiled. Suggest an explanation for these observations.
Lithium hydride is the most stable of the alkali metal hydrides with respect to decomposition into the metal and hydrogen gas, and the decomposition temperatures of the alkali metal hydrides tend to decrease down Group I. (a) Explain how this is consistent with the trend in AH: (MH, s) values in Table 25.2 for lithium, sodium and caesium. (b) How and why does the trend in AH: (MI, s) values differ? Table 25.2 Values of AH: (MH, s) and AH? (MI, s) for the hydrides and iodides of lithium, sodium and caesium
It is possible to make compounds of the anion C1F4- by heating an alkali metal fluoride with chlorine trifluoride: (25.16) MF(s) + ClF3(g) = MClF,(s) However, the reaction does not occur with all the alkali metal fluorides. Which fluoride do you think would be most suitable, and why?
169
So far, we have denoted metal ions in aqueous solution simply by a parenthesized ‘aq’ after the formula of the ion, as in Na+(aq), Mg2+(aq)and A13+(aq),etc. We have now reached a point where it will be useful to look at the interaction between the metal ion and the surrounding water molecules in a more detailed way. This will lead us to a new concept of great importance in modern inorganic chemistry: the concept of a metal complex. Although this Book is concerned with the chemistry of typical elements, such as the alkali metals and alkaline earth metals, it is convenient to introduce this new subject by using a transition element. This is because, in transition-metal chemistry, some of the important changes that we shall be discussing are often marked by changes in colour.
Copper sulfate is usually sold as CuS04.5H20.This blue solid (Figure 26.1) is called a hydrate because, as the formula implies, there are discrete water molecules distributed throughout the structure. In terms of an ionic model, therefore, the structure contains three types of chemical species: Cu2+ions, S042- ions and water molecules. The arrangement revealed by X-ray diffraction is shown in Figure 26.2. Read the caption carefully.
Figure 26.1 Copper sulfate pentahydrate, CuS04.5H20(left), and its blue solution in water (right). 170
Figure 26.2 Structural features of CuS04.5H20.Four of the five water molecules in each formula unit are arranged around the copper ion at the corners of a square, which is picked out in blue. If we take the copper to be a +2 ion, the formula of the square unit is C U ( H ~ O ) Above ~ ~ + . and below the squares are oxygen atoms of sulfate ions through which the C U ( H ~ O )units ~ ~ +are linked into chains. The fifth water molecule in each formula unit is not bound to copper: it binds adjacent chains together by means of the hydrogen bonds that it forms with sulfate oxygens.
The structure can be broken down into chains made up of alternate C U ( H ~ O )and ~~+ S042-ions. The chains are held together by hydrogen bonds formed through one of the five water molecules in each formula unit. Let’s try to justify this breakdown further. In Figure 26.2, lines emanating from the copper ion show its bonding with surrounding atoms. How many, and of what type are these surrounding atoms? Six lines emanate from each copper ion and each one links it to an oxygen atom. Four of the six surrounding oxygen atoms belong to water molecules; the other two belong to sulfate ions. Figure 26.2 also tries to show that the arrangement of the six oxygens around the copper is octahedral. To clarify this, the octahedron is shown enlarged in Figure 26.3, with the copper-oxygen bond lengths included. Figure 26.3 suggests that some of the oxygens are more closely and tightly bound to the Cu2+ion than others. How many of these closely bound oxygens are there, what is their type, and how are they arranged around the copper? ion The four oxygens of the four water molecules are 40 pm closer to the cu2+ than the two sulfate oxygens; the four are arranged at the corners of a square, which is picked out in blue in Figure 26.2.
Figure 26.3 The Cu-0 distances in the octahedron of oxygen atoms around the copper in CuS04.5H20.Four of these oxygens belong to water molecules; the other two to sulfate ions. 171
This especially close binding between Cu2+and four water molecules is further justification for regarding them as a separate unit. It suggests a new way of looking at CuS04.5H20:it contains ions with the formula [ C U ( H ~ O ) ~in ] ~which + , a Cu2+ion is bound to four waters at the corners of a square in square-planar coordination as in Structure 26.1. Thus, instead of saying that the solid consists of one Cu2+ion and five water molecules for every S042-ion, we say that it contains one [ C U ( H ~ O ) ~ ] ~ + ion, and one water molecule for every sulfate ion, The entity [ C U ( H ~ O ) ~is] known ~+ as a complex, or, because it carries a charge in this case, as a complex ion. Note the use of square brackets enclosing the formula of the complex, the charge being placed outside them. Some confirmation that this is a sensible way of looking at CuS04.5H20comes when we dissolve the solid in water. The blue solid gives a blue solution of similar hue (Figure 26.1). We can explain the common blue coloration by arguing that it is a property of the complex ion [ C U ( H ~ O ) ~which ] ~ + , persists when the solid dissolves in aqueous solution. Up till now, we would have written the dissolution of CuS04.5H20in water as: (26.1) CuS04.5H20(s)= Cu2+(aq)+ S042-(aq) + 5H20(1) Write an alternative equation for this change incorporating our explanation of the similar colours of CuS04.5H20(s)and its aqueous solutions. The equation is: [Cu(H20)4]S04.H20(s)= [ C ~ ( H ~ o ) ~ ] ~ ++(S042-(aq) aq) + H20(1)
(26.2)
Notice that we still write ‘(aq)’ after the formula of the complex ion in aqueous solution. Although four water molecules are bound especially closely by the Cu2+ ion, the resulting complex, [ C U ( H ~ O ) ~still ] ~ +interacts , with surrounding solvent water molecules that are further away from the central copper. Another sign of the value of these ideas is revealed when an aqueous solution of ammonia, NH3(aq), is added to our solution of copper sulfate. The light blue solution turns an intense deep violet (Figure 26.4), showing that some new complex has been formed. What happens is that each water molecule at the corner of the square in Structure 26.1 is replaced by an ammonia molecule: H2Y
H20 -CU- OH2 I H20 light blue
NH3
2+
I
H3N- CU-NH~
I
NH3 deep violet
Figure 26.4 The intense deep violet colour of the complex ion [Cu(NH3),I2+(aq), right, contrasted with the blue of [Cu(H20),l2+(aq),left. 172
By recognizing the existence of complexes, we obtain a simple but elegant description of the reaction: the copper ion retains its preference for square-planar coordination, but changes the four atoms to which it is bound from oxygen to nitrogen, by exchanging water for ammonia. Ammonia is not the only molecule that can effect a substitution of nitrogen for oxygen around Cu2+,An especially interesting case occurs when an aqueous solution of the compound commonly called ethylenediamine (Structure 26.2) is added to an aqueous solution of copper sulfate. This molecule contains two nitrogen atoms, and its shape and size are such that they can comfortably occupy two adjacent corners of the square-planar coordination around a Cu2+ion in a complex. Assuming that Cu2+retains its preference for square-planar coordination, and that ethylenediamine reacts with [ C U ( H ~ O ) ~to] ~form + a new complex, how many ethylenediamine molecules will react with each [CU(H,O)~]~+? Two; there are four positions to be occupied around the copper ion, and each ethylenediamine takes care of two of them:
The complex that is formed (Structure 26.3) again has a deep blue colour. Ethylenediamine is often abbreviated to ‘en’; the complex can then be drawn as in Structure 26.4. r
12+
26.4
In this Section we look more closely at the components of a complex, and the way in which they are bound together. In the cases we shall consider, there is a central atom or ion such as Cu2+,and this is bound to and surrounded by groups such as H20, NH3 or en. These groups are called ligands. Each ligand has one or more coordinating atoms, which are the atoms that are bound to the central metal ion. In the case of water this is oxygen; in the case of ammonia, it is nitrogen. Ligands may be neutral molecules or ions. For example, the halide ions, F-, C1-, Br- and I-, act as ligands in many metal complexes. If concentrated hydrochloric acid is added to an aqueous solution containing copper ions, the four water molecules in [ C U ( H ~ O ) ~are ] ~replaced + by chloride ligands. Write an equation for the reaction, making sure that the charge is balanced. The equation is: = [CuCl4l2-(aq) + 4H20(1) [ c ~ ( H ~ O ) ~ ] ~ ++(4Cl-(aq) aq)
(26.5) 173
Note how the substitution of charged ligands for neutral ones alters the overall charge of the complex: from an ionic standpoint, [CuCl4I2- is an assembly of one Cu2+and four C1- ions, making the overall charge -2. Now let us turn to the bonding between the metal ion and the ligands. The ligands that we have introduced are in states in which all the atoms have noble gas configurations. This is clear from Figure 26.5, which shows Lewis structures for three typical examples: H20, NH3 and F-. Notice first that the coordinating atoms in the ligands are all electronegative (0,N and F).
Figure 26.5 Lewis structures for the ligands NH3, H 2 0 and F-; by convention, the electrons, whether bonding or non-bonding, are grouped in pairs.
In Figure 26.5, what else do the coordinating atoms have in common? They all carry non-bonded electron pairs that are are not involved in chemical bonding. For example, nitrogen in NH3 has one non-bonded pair, and oxygen in H 2 0 has two. These observations suggest two crude but useful ways of looking at the bonding between the metal atom, or ion, and the ligand. Firstly, there is an electrostatic viewpoint: the central metal atom carries a net positive charge, so the ligands orientate themselves with the negative charge of the non-bonded pairs directed towards this positive site; the electrical interaction between them binds the metallic element and the ligand together. Secondly, there is a covalent viewpoint: the ligands’ non-bonded pairs become electron pair bonds if the ligands donate them to the metallic element. Ligands such as H20, NH3 and F-, which have just one coordinating atom -one point of attachment through which they can bind to a metal atom or ion -are called unidentate ligands. What do you think a ligand like ethylenediamine is called? It contains two nitrogen atoms with non-bonded electron pairs -two coordinating atoms -and is called a bidentate ligand. There are also polydentate ligands, which contain more than two coordinating atoms. Table 26.1 includes diethylenetriamine (dien), which is tridentate, and the hexadentate ethylenediaminetetraacetate anion (edta4-). In this latter case, the coordinating atoms are the two nitrogens, and four negatively charged oxygen atoms of the four -CH2COO- groups. The geometry of edta4- is such that these six atoms, with their non-bonded pairs of electrons, can occupy all six positions around a metallic element in an octahedral complex (see Figure 28.19).
174
Table 26.1 A few bidentate and polydentate ligands
* Notice that chemists who work with metal complexes often use names and abbreviations for ligands that pre-date modern, systematic organic nomenclature; for example, ethylenediamine (en) has the systematic name 1.2-diaminoethane.
Metallic elements are often found in a combined state as complexes. A complex usually consists of a central metal atom or ion bound to surrounding molecules or ions called ligands. Common ligands include halide ions, water, and ammonia. Ligands are linked to the metal through coordinating atoms, such as oxygen, nitrogen or halogen, which are electronegative and carry non-bonded electron pairs. A ligand may contain more than one coordinating atom. For example, en, dien and edta4- contain two, three and six coordinating atoms, respectively, and are said to be bidentate, tridentate and hexadentate, respectively.
Magnesium sulfate occurs as a colourless hydrate, MgS04.7H20, often called Epsom salts. Part of the structure is shown in Figure 26.6, Study the key carefully, and then answer the following questions. The distances between the magnesium ion and its immediate neighbours are nearly identical. Write the formula of the magnesium complex that this solid contains. How do you think the ligands are arranged around the magnesium ion? Assuming that the complex persists in the aqueous solution when MgS04.7H20dissolves in water, write an equation for the dissolving process. A new complex is formed when this aqueous solution is treated with an aqueous solution of edta4-. Use Table 26.1, and your answer to part (a) of this question, to write an equation for the reaction that occurs.
Figure 26.6 Structural features of MgS04.7H20.
175
Until quite recently, alkali metals were not thought of as elements whose cations formed clearly defined complexes that would persist in a sequence of chemical reactions. In particular, it was not easy to find evidence for alkali metal complexes, either in solution or in the solid state, in which the cation was bound to ligands that were not shared with other cations. For example, when sodium sulfate is crystallized from water, it does so as the hydrate Na2S04.10H20.In this hydrate, each sodium ion is surrounded by six water molecules at the corners of an octahedron, but one cannot describe this unit as a [Na(H20)6]+complex because four of the six water molecules are shared with, and bound to, other sodium ions. This situation changed during the late 1960s with the discovery by Charles Pedersen (Figure 27.1) of ligands known as crown ethers. Typical examples of crown ethers are shown as Structures 27.1 and 27.2.
CH2
H2 I O H/c \ \C CH2
I
I
27.1 dicyclohexyl-18-crown-6
0 HC
C / \
I
C’
0
27.2
CH2
H HC \
I
0
\C
I
CH
II
dibenzo-18-crown-6
Which sites in Structure 27.1 are likely to enable it to act as a ligand, and how many of them are there? The six oxygen atoms carry non-bonded electron pairs like those on the oxygen atom in a water molecule. The molecule is potentially a hexadentate ligand. 176
Figure 27.1 Charles Pedersen ( 1 9 0 6 1989), the son of a Norwegian seaman and mining engineer, was born at Pusan, Korea and went to school in Japan. At the age of 17, he travelled alone to the United States where he took degrees at the University of Dayton, Ohio and the Massachusetts Institute of Technology. In 1927, he joined DuPont, and in the 34thof the 42 years he spent with this company, he obtained some unexpected white needle-like crystals while worlung on organic catalysts. Instead of throwing them away, he studied their properties, and found that in their presence, sodium hydroxide would dissolve in organic solvents such as diethyl ether (Structure 27.3) or benzene (Structure 27.4).The crystals turned out to be those of the crown ether 27.2, which he named dibenzo18-crown-6 because the central ring bridges two benzene rings and includes eighteen atoms, of which six are oxygen. For the discovery of such crown ethers and their properties, Pedersen shared the 1987 Nobel Prize for Chemistry.
Pedersen’s discovery of crown ethers was itself accidental, but it was a second accidental observation that ultimately led to the award of the Nobel Prize for Chemistry. As you know, it is an accepted property of ionic compounds, such as alkali metal salts, that they are almost insoluble in non-polar organic solvents such as heptane, trichloromethane (chloroform) and benzene. In other words, if we try to dissolve potassium iodide in chloroform: (27.1) KI(s) K+(org) + I-(org) we find that the equilibrium lies well over to the left. Pedersen found that, in many cases, this situation was reversed when a crown ether was added to the organic solvent. For example, in the case that we have just discussed, in the presence of the molecule shown as Structure 27.1, potassium iodide will dissolve in chloroform to the tune of 7 grams per litre.
How can you explain this by using Le Chatelier’s principle? If the cyclic ether acts as a ligand and complexes with the alkali metal cation, it will disturb the equilibrium in Equation 27.1, shift it to the right, and increase the solubility of KI in chloroform. This capacity of crown ethers to make alkali metal salts soluble in organic solvents has proved extremely useful in organic chemistry. For example, potassium permanganate (KMn04), contains the permanganate ion, Mn04-, which will oxidize many organic compounds. However, there are difficulties in bringing the reactants into intimate contact. Solid KMn04 is likely to react slowly. In trying to bring the reactants together in a common solvent, one finds that organic solvents that dissolve the organic reactant will not dissolve KMn04, and a solvent like water will dissolve KMn04, but may not dissolve the organic reactant, or may sometimes even react with it. These difficulties can be circumvented by using benzene (27.4) as the organic solvent, and adding both KMn04 and the cyclic ether 27.1. The solubility of KMn04 in benzene is then as much as 30 g litre-]. Many organic compounds can be dissolved in this benzene solution, and oxidized smoothly and easily at room temperature.
H 27.4
Solid compounds containing alkali metal complexes with cyclic ether ligands can be crystallized from a solution of an alkali metal salt and a cyclic ether in an organic solvent. For example, Figure 27.2 shows part of the crystal structure of a compound obtained from a solution of sodium thiocyanate, NaNCS, and the crown ether 27.2 in methanol. The sodium ion sits just below, but close to the centre of the inner ring of six oxygens on the crown ether ligand. As expected then, the ligand is hexadentate, and we can think of the complex as [NaL]+,where L is Structure 27.2. The other close neighbour of the sodium is the nitrogen of the thiocyanate ion, NCS-, which is some 50 pm further away than the oxygens of the ether. The compound is often said to contain ion pairs, [NaL]+[NCS]-, in which the cation is the alkali metal complex. A measure of the stability of such alkali metal complexes in solution is the equilibrium constant of the reaction M+(solv) + L(so1v) = [ML]+(solv)
(27.2)
where ‘solv’ denotes some general solvent. When L is Structure 27.1, and the solvent is methanol, the potassium complex (M = K) is the most stable as
177
Figure 27.3 shows. Now, Figure 27.2 suggests that the alkali metal cation in these complexes sits close to the centre of the encircling ligand. Perhaps then, the potassium complex is the most stable because the potassium ion fits most snugly into the cavity in the cyclic ether. One way of testing this hypothesis is to decrease the size of the cavity by using ligand 27.5, whose inner ring contains two fewer carbon atoms and two fewer oxygen atoms. When this change is made, potassium no longer forms the most stable complex. Would you expect sodium or rubidium to form the more-stable complex? Sodium; the cavity becomes smaller when the ligand change is made. As Na+ is smaller than K+ or Rb+, it is this ion that will now fit more snugly into it.
Figure 27.2 Structural features of the compound [NaL]+NCS-, where L is dibenzo- 18crown-6 (27.2); the hydrogen atoms have been omitted for clarity. Sodium is coordinated by the six oxygen atoms of the crown ether, the Na-0 distances being about 280 pm. The distance between sodium and the nitrogen atoms in the thiocyanate ion, -N=C=S, is 332 pm. The diagram shows why such ligands are called crown ethers: the ligand has a crown-like shape and the sodium is topped by the cavity of the crown.
Figure 27.3 The equilibrium constant of Reaction 27.2 for the cases where M is an alkali metal, the solvent is methanol, and the ligand L is dicyclohexyl- 18-crown-6 (27.1). Note that the scale on the vertical axis is labelled in powers of 10. 178
H2
C
27.5 dicyclohexyl-14-crown-4
Notice, however, that the ‘cavity’ in crown ethers, such as 27.1,27.2 and 27.5, is a cavity in a ring and is therefore close to two dimensional. Soon after Pedersen published his work on crown ethers, the French chemist, Jean-Marie Lehn (Figure 27.4), set out to modify the cyclic ether structures so that the cavity became three dimensional. The best known of these new ligands is shown as Structure 27.6.
27.6
How many coordinating atoms does Structure 27.6 contain? There are eight: the six oxygens and two nitrogens all carry non-bonding electron pairs. This ligand contains three rings with potential bonding sites, and so the cavity at its centre is three dimensional. The ligand and its bonding atoms can wrap themselves around a cation of the appropriate size, forming a ‘crypt’, which conceals and protects the cation from surrounding molecules or ions. For this reason, such ligands are known as cryptands; Structure 27.6 is called cryptand-222 and written C222.The three twos refer to the three pairs of oxygen atoms which lie on the three chains that link the nitrogen atoms. One effect of this new arrangement is that the complexes become much more stable. For example, the equilibrium constant of the reaction K+(methanol) + C222(methanol)= [KC222]+(methanol) (27.3) is about 1O’O mo1-I litre, about 10 000 times the value with crown ether 27.3 plotted in Figure 27.3. In Figure 27.5, we plot the equilibrium constants for reactions like Equilibrium 27.3 for each of the alkali metal cations; as in Figure 27.3, there is a maximum at potassium. However, sodium ions also form very stable complexes with C222,and, as you will now see, this led to a striking discovery in alkali metal chemistry.
Figure 27.4 Jean-Marie Lehn, Professor of Organic Chemistry at the University of Strasbourg. He developed ligands such as cryptands, which contain three-dimensional cavities. With Charles Pedersen (Figure 27.1), he shared the 1987 Nobel Prize for Chemistry for contributions to the idea of molecular recognition. Figure 27.5 provides a very simple example of this idea. Cryptand-222 (27.6) forms a complex with potassium whose equilibrium constant (Equation 27.2) is some 100 times greater than that with any other alkali metal cation. The ligand’s size and shape are such that it will ‘recognize’ and preferentially combine with one of a number of otherwise quite similar species. 179
Figure 27.5 The equilibrium constants of Reaction 27.2 for the cases where M is an alkali metal, the solvent is methanol, and the ligand is cryptand-222 (27.6).
In Section 23.1.2, you saw that all five alkali metals dissolve in liquid ammonia to give blue solutions containing ammoniated electrons: M(s) = M+(amm) + e-(amm)
(27.4)
Since the blue colour, and the absorption spectrum that goes with it (Figure 27.6), is a property of the ammoniated electrons and not of the alkali metals, it is common to solutions of all five alkali metal-ammonia complexes.
Figure 27.6 The absorption spectrum of the blue solutions obtained by dissolving small amounts of alkali metal in liquid ammonia. The absorption is attributed to the ammoniated electron, e-(amm).
However, the alkali metals will also dissolve in other solvents that have some resemblance to liquid ammonia. One of these is aminoethane, C2H5NH2,which is a liquid at room temperature. Again the solutions have a blue colour, but the spectra for the different metals are quite different: they are shown in Figure 27.7. The absorption peak for solvated electrons in ammonia and related solvents usually occurs in the wavelength range marked by the pink band in Figure 27.7. Which metal most clearly gives solvated electrons? Only lithium gives an absorption peak in this region,
180
Figure 27.7 The absorption spectra of the solutions obi:ained when the alkali metals dissolve in aminoethane.
Because sodium, potassium, rubidium and caesium give absorption bands with peaks at distinctly different wavelengths, these bands must be generated by species that contain the different metallic elements. Look at Figure 27.6: is it likely that these bands are a property of the solvated cations, M+? Not very likely; no such bands appear in the spectra of alkali metal solutions in liquid ammonia, which contain M+(amm): the cations appear not to absorb at these wavelengths. This and other evidence led to the suggestion that, in aminoethane (ameth), the dissolution of the alkali metals takes place chiefly by the route 2M(s) 6 M+(ameth) + M-(ameth) (27.5) The peaks in the absorption spectra of Na, K, Rb and Cs in aminoethane solution (Figure 27.7) can then be attributed to solvated alkali metal anions. This is not quite as outlandish as it might appear: in Table 23.1, we quoted the electron affinities of the alkali metals, and they were positive. This shows that the uptake of electrons by gaseous alkali metal atoms is exothermic; for example Na(g) + e-(g) = Na-(g); AH: = -53 kJ mol-l (27.6) The sign of the energy change for this step at least, is favourable to the formation of compounds or solutions containing such anions. This is not the case for the noble gases or the Group I1 elements, which precede and follow, respectively, the alkali metals in the Periodic Table. A possible reason for this is provided by the change in electronic configuration that occurs during a reaction like Equation 27.6. What is the electronic configuration of the alkali metal anion Na-? ls22s22p63s2:in Reaction 27.6 an electron is added to the outer nsl configuration of the metal atom to form a full ns2 sub-shell.
181
In one sense then, the acquisition of an electron by a sodium atom when forming Na- resembles the acquisition of an electron by a chlorine atom when forming C1-: in both cases, a sub-shell is filled (3s in the case of sodium; 3p in the case of chlorine). Of course, in the chlorine case, the filling of the sub-shell completes the stable noble-gas configuration, a fact reflected in the much larger electron affinity (349 kJ mol-l). Nevertheless, the analogy, combined with the still fairly favourable AH: of Reaction 27.6, makes the idea of compounds containing alkali metal anions more plausible than it might first seem. How can we prepare such a compound? You have seen that when sodium is added to aminoethane until no more will dissolve, the following equilibrium is set up: 2Na(s) +Na+(ameth) + Na-(ameth) (27.7) Unfortunately, very little sodium will dissolve in aminoethane: the equilibrium lies well over to the left. One sign of this is that if we try to make compounds of the type Na+Na- or K+ K- by evaporation of solutions of alkali metals in aminoethane, we obtain the original metals rather than compounds containing anions. How might we shift the equilibrium in Equation 27.7 to the right, and so stabilize the solution containing Na-? A crown ether or cryptand that complexes with Na+ might do the trick. In the late 1970s, Professor James Dye of Michigan State University used C222for this purpose. In the presence of this ligand, sodium dissolved in aminoethane to form [NaC222]+,and the solubility of the metal was increased to around 5 g litre-l. The equilibrium can now be written: 2Na(s) + C222(ameth)T [NaC2,,]+(arneth) + Na-(ameth) (27.8) When the solution was cooled to -15 OC, shiny golden crystals of the compound [NaC222]+ Na- were precipitated (Figure 27.8). The compound did not decompose below 0 "C provided it was rigorously protected from air and moisture by keeping it in a dry atmosphere of pure nitrogen or argon. Above 0 O C , it tended to decompose to sodium metal: (27.9) [NaC222]+ Na- = 2Na(s) + C222(s)
Figure 27.8 Gold-coloured crystals of the compound formulated [NaC222]+Na-, which contains an Na- ion.
The crystal structure of the complex is shown in Figure 27.9. As you can see, the sodium in the complex cation is surrounded by the C222ligand and bonded to all eight of its donor atoms. Although it is quite difficult to spot, the Na- ions are arranged around the complex cation at the corners of an octahedron: in the Figure, the Na- ions have been connected by triangles that form the faces of the octahedron. 182
Figure 27.9 The structure of [NaC222]+Na-. The sodium at the centre lies within the cavity of the cryptand-222, where it is coordinated by six oxygen and two nitrogen atoms. Around the [NaC222]+complex, there are six Naions at the corners of an octahedron.
1
2
3 4
5
The alkali metal cations form relatively few discrete complexes in solution. However, during the 1960s and 1970s, they were found to form complex ions with polycyclic ethers such as dicyclohexyl- 1 8-crown-6 (27.1) and cryptand222 (27.6). If a crown ether is added to benzene, the benzene then dissolves some common alkali metal salts, such as KMn04, because the complexing between the alkali metal cation and the crown ether makes the otherwise unfavourable dissolution reaction favourable. The resulting solution of Mn04- in benzene can be used as an oxidizing reagent in organic chemistry. Such complexes have their greatest stability when the size of the alkali metal cation and the size of the cavity in the ligand are suitably matched. In liquid ammonia the alkali metals form solvated cations, M+, and solvated electrons; in aminoethane, the major products with sodium, potassium, rubidium and caesium are solvated cations, M+, and solvated anions, M-. However, the solubility of the metals in this latter case is very small. The addition of cryptand-222 to the aminoethane and the resulting complexing with the cations, M+, greatly increases the solubility of the alkali metal. At -15 "C, the sodium solution then deposits crystals of the salt [NaC222]+ Na-, which contains a sodium anion.
In cryptand-222 (27.6), the two nitrogen atoms are linked by three chains of structure -CH2-CH2-0-CH2-CH2-0-CH2-CH2-. In cryptand-221, one of these three chains is shortened to -CH2-CH2-0-CH2-CH2-, and in cryptand-2 11, two of the three chains are so shortened. With cryptand-222 in methanol, it is potassium that forms the most stable alkali metal complex. With cryptand-221, it is a different alkali metal cation, and with cryptand-21 1, a different cation again. Identify the alkali metal cations that form the most stable complexes in methanol with cryptand-221 and with cryptand-21 1. 183
In the final Section of the main text, we discuss the chemistry of the elements beryllium, magnesium, calcium, strontium, barium and radium. They occur in Group I1 of the Periodic Table (Figure 28.1), and are often called the alkaline earth elements. The word ‘earth’ is an old-fashioned name for metal oxides of very high melting temperature, which can be obtained by subjecting minerals containing the elements to heat or some other simple process. When the oxides MgO, CaO, SrO, BaO and RaO (but not BeO) are added to water, they form hydroxides that are soluble enough to give solutions that are alkaline (hence the term aLkaLine earth eLement): (28.1) MO(s) + H20(1) = M(OH)~(S) M(OH)2(s) = M2+(aq)+ 20H-(aq) (28.2) Of the six elements, calcium and magnesium are by far the most common, being the fifth and sixth most abundant elements in the Earth’s crust. Both elements occur as dolomite, CaMg(C03)2 (Figure 28.2), which is a major constituent of certain mountain ranges, such as the Dolomite Alps in Italy. Magnesium occurs by itself in deposits of magnesite, MgC03, and after sodium, it is the most abundant metallic element in seawater, of which it constitutes about 0.13% by mass.
Figure 28*1 The Group I1 elements.
The most important sources of calcium are the huge sedimentary deposits of calcium carbonate, CaC03, which are formed from the fossilized remains of longdead shellfish. In this category come the minerals limestone, chalk and marble. Dissolved carbon dioxide makes rainwater slightly acid; consequently, limestone is slightly soluble in rain: CaC03(s) + H+(aq) = Ca2+(aq)+ HC03-(aq)
(28.3)
The resulting erosion gives rise, even at moderate heights, to rugged landscapes of a quality that in other rocks are confined to coastlines and mountains (Figure 28.3). Hard limestone country is often characterized by inland cliffs, caves, gorges and springs. Cheddar Gorge in Somerset and Ingleborough in Yorkshire are fine British examples. The erosion arising from Reaction 28.3 also develops open pores, cracks and fissures in limestone strata so that they become excellent aquifers (media for the storage and transmission of water). In Britain, limestones are the source of more drinking water than all other aquifers put together. Figure 28.4 shows how limestone strata warm up water from the Mendip Hills as they carry it some 15 km to Bath, where it emerges in the famous hot springs. The most important beryllium mineral is the aluminosilicate, beryl, Be3A12(Si03)6. Emeralds have the same composition, except that they contain about 2% chromium, which provides the green colour. Both strontium and barium occur naturally as the sulfates celestite, SrS04, and barite, BaS04. Other Group I1 minerals are the carbonates strontianite, SrC03, and witherite, BaC03.
184
Figure 28.2 Large crystals of the mineral dolomite.
Figure 28.3 Hard limestone country, such as Cheddar Gorge, is among the most attractive in the British Isles.
Figure 28.4 Limestone strata transmit water from the Mendip Hills to Bath, warming it up as they do so.
Radium occurs in association with uranium because 2i$ Ra , its most long-lived isotope, is one of the intermediate products in the long chain of radioactive decays that converts 2;!U into 2g$Pb.The isotope 'i;Ra undergoes a-decay: (28.4) "!Ra = 'i;Rn + ;He ; t%= 1600 years About 10 tonnes of uranium ore will yield 1 mg of radium.
The redox potentials of the Group I1 metals, E8(M2+I M), are comparable with those of the alkali metals (Table 21.2). This shows that they are very powerful reducing agents. What method is often used to extract such metals from their ores or compounds? The electrolysis of molten compounds (Section 17.3).
185
All of the metals are extracted by the electrolysis of their molten chlorides, although other compounds such as potassium or sodium chloride are usually added to the melt to lower the melting temperature and improve the conductivity. In industry, magnesium is much the most important of the metals. The magnesium chloride for the electrolysis is obtained from brines of the type mentioned in Section 23, or even from seawater. The Dead Sea, where Israel built new electrolytic plants during the 1990s, is an attractive source (Figure 28.5). A typical electrolytic cell for magnesium production is shown in Figure 28.6.
Figure 28.5 A satellite photograph showing the Dead Sea which is fed from the north by the Jordan River (A). With no outflow, evaporation leads to a concentration of dissolved salts which is six times that of seawater. To the south, region B is divided into salt evaporators, from which magnesium chloride and other useful salts are obtained.
Figure 28.6 The production of magnesium. The melt temperature is about 700 "C. Molten magnesium is liberated at the negative electrodes, and rises to the surface where it is skimmed off and cast into ingots. 186
With a density of only 1.74 g ~ m - magnesium ~, is the lightest structural metal, and this property is retained when it is alloyed with other metals, such as aluminium, to increase its strength and heat resistance. Magnesium alloys are therefore used extensively in aircraft construction. Other uses include lightweight car engine cases, and luggage. In 1999, world production of magnesium was 380 000 tonnes. Some physical properties of the metals, their atoms and their ions are shown in Table 28.1.
Table 28.1 Some properties of Group I1 metals, atoms and ions
Note that in each case, the density, melting temperature and value of AHZm(M, s) are greater than the values for the adjacent alkali metal in the Periodic Table (compare Table 23.2). The very negative values of P ( M 2 +I M) suggest that the alkaline earth metals are reducing agents of a strength comparable with that of the alkali metals. However, kinetic factors often make their reducing properties less spectacular. All of the metals dissolve readily in dilute acids with evolution of hydrogen: (28.5) M(s) + 2H+(aq) = M2+(aq)+ H2(g) The reactions with water are more various. With calcium, strontium and barium, the reaction is brisk and hydrogen is evolved; for example Ca(s)
+ 2H20(1) = Ca2+(aq) + 20H-(aq) + H2(g)
(28.6)
In the case of calcium, the calcium and hydroxide ions soon accumulate to a concentration at which the sparingly soluble calcium hydroxide, Ca(OH)2, is precipitated. The reaction of magnesium with water, however, is very slow and that of beryllium is negligible. This is because these two elements form relatively insoluble oxides and hydroxides which create a protective film on the metal surface. In general, however, the reactions that you have met show that the Group I1 metals are readily oxidized to substances containing ions with a charge of +2. These ions have a noble gas configuration. However, oxidation does not proceed further to substances containing ions with a charge of +3 or more.
187
What term in Table 28.1 is relevant to this failure to observe further oxidation? The third ionization energies, Z3,of the Group I1 elements are enormous. It is extremely difficult to remove a third electron from a Group I1 atom. The very high values of Z3 are most clearly demonstrated by a comparison of the first three ionization energies of magnesium with those of the adjacent Group I11 metal, aluminium (Table 28.2). Table 28.2 Ionization energies of magnesium and aluminium
Why is I3 for a Group I1 atom such as magnesium so large? The third ionization energy, Z3,is the ionization energy of the ion, Mg2+.This ion has the noble gas configuration of the neon atom. On ionization, an electron must be removed from the inner shell (principal quantum number n = 2), whose electrons are much closer to the nuclear charge.
The Group I1 metals have higher melting temperatures, higher enthalpies of atomization and higher densities than their alkali metal neighbours (Figure 28.7) of the same Period. Explain this in terms of the electron gas model of metallic bonding.
Strontium and barium react with water, evolving hydrogen gas. In this they resemble rubidium and caesium. The reactions of the Group I1 elements, however, are much slower than those of the adjacent Group I elements. Suggest a reason for this.
Figure 28.7
Relative positions of the alkali metals, the Group TI elements and aluminium in the Periodic Table.
As noted in Section 28.1, the Group I1 elements cannot form ionic trihalides such as MgF3, because the third ionization energies of these elements are so large. But a comparison of the first ionization energies of, say, potassium and calcium, or rubidium and strontium (Table 28.3) shows no enormous differences. Why then cannot the Group I1 elements form solid monohalides, MX(s), like the alkali metals? To answer this question, we shall compare the cases of KC1 and CaCl in particular detail. We begin with the Born-Haber cycle for a metal chloride, MCl(s) (Figure 28.8). From this, we obtain the equation AHy(MC1,
188
S)
= M Z m ( M ,S)
+ Zl(M) +
$D(Cl-Cl) - E(C1) + L(MC1, S)
(28.7)
Table 28.3 First ionization energies of some Group I and Group I1 elements
Figure 28.8 The Born-Haber cycle for a metal chloride MCl(s).
Table 28.4 shows some of the quantities that must be used in this equation when it is applied to KCl and CaC1. Table 28.4 Terms in the Born-Haber cycles for potassium and calcium monohalides
Use Equation 28.7 and the data in Table 28.4 to calculate the lattice energy of KC1, L(KC1, s). L(KC1, s) = -718 kJ mol-'; the exercise is the one carried out on p. 124: the top five numbers in column 2 must add up to the bottom number. As you can see, there is no value of M y ( C a C 1 , s) in Table 28.4. This is because the compound has never been made, so its enthalpy of formation has not been determined experimentally. Let's try to obtain a value by other means. If we think in terms of ionic compounds, KCl(s) contains K+ and C1- ions, and CaCl(s) would contain Ca+ and Cl- ions. Calcium is adjacent to potassium in the Periodic Table, and the ion Ca+ has the same charge, and only one more proton and one more electron than K+. Let us therefore assume that the two ions have similar sizes.
189
Now, the lattice energy of an ionic solid is given approximately by the Kapustinskii equation: (20.11) L = - wvz+zr+ + rFor both KCl and CaC1, v = 2,Z+ = 1 , Z = 1 and Y- = r(Cl-). We have also assumed that r(Ca+) = r(K+). What does this suggest about the lattice energies of CaCl(s) and KCl(s)? They would be identical: L(CaC1, s) = L(KC1, s) = -7 18 kJ mol-l. You can therefore put this value of L(CaC1, s) into the last column of Table 28.4, and combine it with the other data to obtain a value of AHy(CaC1, s). Do this now. AHy(CaC1, s) = -177 kJmol-', the sum of the top five numbers in the column. The fact that this is a negative value is extremely interesting, It suggests that CaCl(s) might well be stable with respect to the elements of which it is composed. Thus, the reverse of the formation reaction, the decomposition into calcium and chlorine, is endothermic: CaCl(s) = Ca(s) + ;C12(g); AH: = 177 kJ mol-' (28.8)
To check whether CaCl(s) is thermodynamically stable to this reaction, we need the S : as well as that of AH.: As the entropy changes for analogous value of A reactions are similar (Section 16), AS: can be estimated by using the value for the analogous reaction (28.9) KCl(s) = K(s) + Cl&) From the Data Book, AS: = P ( K , S) + iP(C12, g) - Se(KC1, S) = 93.2 J K-I mol-l Using this as an estimate of AS:
AGE = AH:
for Reaction 28.8,
TAS: = 177 kJ mo1-l - (298.15 K x 93.2 J K-' mol-') = 177 kJ m o t 1 - 28 kJ mol-' = 149 kJ mol-I -
(28.10)
This tells us that CaCl(s) is thermodynamically stable with respect to its constituent elements at 298.15 K, and that AGF(CaC1, s) = -149 kJ mol-I. In other words, when we mix calcium metal and chlorine at room temperature, the formation of CaCl(s) is thermodynamically favourable. So why has no one succeeded in making it? One possibility is that the reaction hasn't been properly tried out, and that we may be able to do better! But before we try the experiment, there is another possibility we should think about. It may be that although CaCl(s) is stable with respect to its elements, it is unstable to some other sort of decomposition. Suggest such a decomposition.
190
Since calcium so readily forms a dichloride, CaC12,one possibility is 2CaCl(s) = Ca(s)
+ CaC12(s)
(28.11)
In this reaction, some of the Ca+ ions in CaCl undergo reduction to calcium metal, and others are oxidized to Ca2+ions in CaC12.This process of simultaneous oxidation and reduction is called disproportionation. Use our estimated value of AGy(CaC1, s) and the Data Book value AG?(CaCl2, s) = -748 kJ mol-' to estimate AG: for Reaction 28.11. AG:
= AG7(CaCl2, s) - 2AGF(CaCl, s) = -748 kJ mol-1 - 2 x (-149 kJ mol-')
= -450 kJ mol-'
Thus, although solid Group I1 monohalides are thermodynamically stable with respect to their elements at 25 OC, they are thermodynamically unstable with respect to disproportionation into the metal and MC12, which contains an M2+ion with the noble gas configuration. This is the reason why such compounds do not exist.
Since the ion Ca+ contains one more electron than K+, you may be unhappy with the assumptions used in Section 28.2, namely that the ionic radii of K+ and Ca+, and therefore the lattice energies of KCl and CaCl, are of similar sizes. Suppose instead, therefore, that Ca+ has the same radius as Rb+, the next largest alkali metal cation. Given that the lattice energy of RbCl is -692 kJ mol-', reassess the stability of CaCl(s) using the method of Section 28.2. Does the change affect the conclusions in any serious way?
The alkali metals do not form dihalides. Table 28.5 contains the thermodynamic data that are needed to calculate the lattice energy of BaF,(s). By assuming that the ionic radii of Ba2+and Cs2+are identical, estimate AHy(CsF2, s). Is your value consistent with the difficulties experienced in trying to make C S F ~You ? should refer to the Data Book in answering this question. Table 28.5 Terms in the Born-Haber cycles for barium and caesium difluorides
191
Lime is a general term, which includes quarried calcium carbonate, CaC03, and two other compounds that are easily obtained from it: quicklime or calcium oxide, CaO, and slaked lime or calcium hydroxide, Ca(OH)2. When chalk or limestone is roasted at about 1 200-1 400 "C (Figure 28.9), it loses carbon dioxide, leaving quicklime as a white powder: CaC03(s) = CaO(s) + C02(g) (28.12)
Figure 28.9 The chalk-bearing countryside of England is scattered with the remains of old limekilns. This picture shows the brick pot in which alternate layers of chalk and burning coke produced quicklime.
If water is added to quicklime, a vigorous reaction occurs with evolution of heat: the product is slaked lime: CaO(s)
+ H20(1) = Ca(OH)2(s)
(28.13)
It is also customary to speak of dolomitic lime, which describes dolomite, CaMg(C03)2,and the products obtained by treating this substance in the same way as CaC03. Lime has the power to neutralize acidic substances. Ca(OH)2, for example, is sparingly soluble in water (1.3 g litre-' at 25 "C), and so both Ca(OH)2 and -because of Equation 28.13 -CaO give alkaline solutions in water: Ca(OH)2(s) = Ca2+(aq)+ 20H-(aq) (28.14) The hydroxide ions can then neutralize H+(aq): H+(aq) + OH-(aq) = H20(1)
(28.15)
The corresponding neutralization reaction of CaC03 was given at the beginning of Section 28: CaC03(s) + H+(aq) = Ca2+(aq)+ HC03-(aq) (28.3) This equation, however, is appropriate only at moderate acidities, such as those found in rainwater (around pH 5 ) . If the acidity is high, the hydrogen carbonate ion reacts further: HC03-(aq) + H+(aq) = H20(1) + C02(g) (28.16) ~
Write the overall reaction for the neutralizing effect of CaC03(s) in high acidities. It is the sum of Equations 28.3 and 28.16, which is: CaC03(s) + 2H+(aq) = Ca2+(aq) + C02(g) + H20(1) 192
(28.17)
Very large amounts of lime are consumed in neutralizing acidity. Pulverized limestone, slaked lime and quicklime are all used, where necessary, to lower the acidity of soils. Lime is also used to alleviate the effects of acid rain, which is caused by atmospheric reactions of the sulfur dioxide and oxides of nitrogen produced in coal or oil-fired power stations; for example S02(g) + H20(1) + ;O,(g) = 2H+(aq) + S042-(aq)
(28.18)
You can learn more about this use of lime in the Case Study 'Acid Rain: Sulfur and Power Generation' in Elements of the p Block 3. The chemical basis of other important uses of lime is revealed by regarding Reaction 28.12 as an equilibrium system: (28.12) CaC03(s) +CaO(s) + C02(g) The equilibrium lies well to the right at about 1 200 "C; CO&) is driven off, and at this temperature CaO( s) is the solid remaining. However, at ordinary temperatures, the equilibrium lies to the left, and both CaO(s) and Ca(OH)2(s) will combine with C02(g) to form CaCO,; for example (28.19) CaO(s) + C02(g) = CaC03(s) In this reaction, the oxide ion from CaO combines with the non-metal oxide C 0 2 to form a new anion, C032-:
r
-9
1
(28.20) Is there any change in the number of bonds formed by carbon in this reaction?
No; the carbon-oxygen double bond in C 0 2 is replaced by two C-0- single bonds. Here then, a non-metallic atom exchanges two bonds formed with atomic oxygen for two bonds formed to oxygens bearing a negative charge. This is a general characteristic of some important reactions of lime. For example, Si02, or sand, is a network of Si-0 single bonds (Figure 28.10). At high temperatures, calcium oxide will combine with it, opening up some of the Si-0-Si linkages, and replacing them with two Si-0- bonds (highlighted in colour in Figure 28.11).
Figure 28.10 The essential features of the structure of silica. SO3. The compound consists of a network of Si-0 single bonds. Each silicon atom forms four of these bonds, which are tetrahedrally disposed. Each oxygen atom forms two, and the LSiOSi bond angle is about I
&
145".
Figure 28.11 A representation of the structural changes that take place when calcium oxide reacts with silica at high temperatures. Each oxide ion attacks the silica structure, converting two Si-0 bonds into two Si-0- bonds.
Again, two bonds formed by the non-metal with oxygen atoms are replaced by two bonds formed to oxygens that bear negative charges.
193
At temperatures above 1 000 "C, the product of the reaction in Figure 28.11 is a melt consisting of Ca2+ions and anions composed of Si-0 networks and Si-0- bonds. This is exploited in the glass industry, which has a very long history (Figure 28.12).
Figure 28.12 The famous Portland vase is made from Roman glass of the early lSt century AD. It is violet-black, overlaid with white glass, into which figures are cut in cameo relief. The Duke of Portland lent it to the British Museum, where it was vandalized and completely shattered in 1845. A remarkably skilful reconstruction has left little trace of this disaster. In 1945, the Museum bought the vase outright.
Most manufactured glass is of the soda-lime variety. Sodium carbonate and limestone are heated above 1 000 "C with sand. At such high temperatures, the carbonates behave like the combinations (Na20 + C02) and (CaO + C02). The carbon dioxide gas leaves the furnace, and the oxide anions open up Si-0- sites on the silica network in reactions of the type shown in Figure 28.11. The result is a liquid in which Na+ and Ca2+ions are coordinated mainly by the 0- sites on Si-0 network anions. On cooling, the liquid becomes progressively more viscous, forming the glass of everyday experience (Box 28.1). A similar reaction to the one described in Box 28.1 occurs in the blast furnace, where lime removes silica and other non-metallic impurities into a liquid slag that floats above the molten iron. It should be clear from these applications alone that lime is a very large tonnage chemical; world consumption of limestone rock amounts to hundreds of millions of tonnes per annum. Like the other alkaline earth metal carbonates, CaC03 has a very low solubility in water. It is, for example, precipitated when solutions containing Ca2+(aq)are treated with sodium carbonate solutions: Ca2+(aq) + C032-(aq) = CaCo3(s) (28.21)
194
The white suspensions of insoluble carbonate, however, will dissolve if carbon dioxide is bubbled through the solution. This is a special case of Equation 28.3, in which C 0 2 acts as the source of acidity: CaC03(s) + H20(1) + C02(g) +Ca2++ 2HC03-(aq) (28.22) If, however, one tries to make solid Ca(HCO& from the resulting clear solution by evaporation, the reactants are reformed because Ca(HC03)2is very unstable to decomposition: Ca(HC03)2(s) = CaC03(s) + CO,(g) + H20(1) (28.23) Equation 28.22 is the key to some important natural phenomena. Because of dissolved atmospheric C02, natural water percolating through limestone can become saturated with dissolved calcium hydrogen carbonate. The dissolved calcium is an important contribution to the hardness of such water. However, boiling eliminates this hardness by returning equilibrium in Reaction 28.22 to the left and destroying the dissolved hydrogen carbonate. A similar reversion to insoluble calcium carbonate may result in the formation of striking stalagmites and stalactites from water droplets in limestone caverns (Figures 28.14 and 28.15).
195
Figure 28.14 The evaporation of dripping water containing dissolved Ca2+(aq)and HCO,-(aq) on surface irregularities in limestone caverns can shift the equilibrium in Reaction 28.22 to the left, and give rise to needle-like structures called stalagmites and stalactites. Stalactites are needles of limestone suspended from the roof; stalagmites rise up from the floor.
0 II
Figure 28.15 Stalagmites and stalactites in a limestone cave.
The highest normal oxide of phosphorus has the empirical formula P205,and contains discrete P4O10 molecules (Structure 28.1). Calcium oxide reacts with this oxide to give calcium phosphate, Ca3(P04)2: 6CaO(s) + P4OIo(s) = 2Ca3(P04)2(~) (28.24) in which the phosphate ion, Po43-, has Structure 28.2. Show how the kind of structural change that occurs when CaO reacts with C 0 2 and Si02, as in Reaction 28.20 and the reaction shown in Figure 28.11, also takes place here. Y 96
28.1
0 II
.P. -0’ -0028.2
If the Group I1 metals are heated in air at 600 OC, beryllium, magnesium, calcium and strontium form normal oxides, MO. Barium forms a peroxide, BaO,; however, if the temperature is increased to 800 OC,the peroxide decomposes to BaO. Beryllium oxide crystallizes with the wurtzite structure (Figure 28.16); the coordination number of beryllium is four. All of the other oxides have the NaCl structure, in which the alkaline earth metal ion is six coordinate.
Figure 28.16 The wurtzite structure of BeO. Each beryllium is tetrahedrally surrounded by four oxygens, and each oxygen is tetrahedrally surrounded by four berylliums.
The hydroxides Be(OH)2, Mg(OH)2 and Ca(OH)2 have very low solubilities in water, and can be precipitated by adding sodium hydroxide to solutions containing the aqueous ions. They will dissolve in hydrochloric acid and nitric acid, but Be(OH)2 alone will also dissolve in strongly aEkaEine solutions, forming the complex [Be(OH),I2-(aq) (Structure 28.3): Be(OH)2(s) + 20H-(aq) = [Be(OH)4]2-(aq) (28.25)
OH I
*-
[ w p H j 28.3
You will recall that oxides or hydroxides that neutralize both acids and bases are called amphoteric. Both the hydroxides and the carbonates decompose to the oxides when heated: (28.26) (28.27)
M(OH),(s) = MOW + H20(g) MCO,(s) = MO(s) + CO2(g) Table 28.6 shows the decomposition temperatures and AH: carbonate reaction. They increase down the Group.
values for the
197
Table 28.6 Values of mz (at 298.15 K) and the decomposition temperature for the decomposition reaction of Group I1 carbonates
Explain why only barium forms a peroxide when the metals Be, Mg, Ca, Sr, and Ba are strongly heated in oxygen.
By using a thermodynamic cycle of the type shown in Figure 25.6, explain why the stability of the alkaline earth metal carbonates increases down the Group.
Magnesium undergoes little or no reaction with water or with sodium hydroxide solution; beryllium does not react with water, but it dissolves readily in sodium hydroxide solution with evolution of hydrogen. Explain this, and write an equation that includes aqueous hydroxide ions for the reaction of beryllium metal with water.
The Group I1 dihalides can be made by the reactions of the metals with the halogens. Up till now, we have treated the bonding in Group I1 compounds and solution as ionic. The dihalides, however, do not altogether fit into this framework. With the exception of MgF2, CaF2, SrF2 and BaF2, which are sparingly soluble, all the dihalides dissolve readily in water to give conducting solutions:
MX2(s) = M2+(aq) + 2X-(aq)
(28.28)
So far, so good. However, in the structure of an ionic solid, each supposed ion should be surrounded by ions of opposite charge. Table 28.7 shows that four of the dihalides have Cd12 or CdC12 layer structures, in which the supposed halide ions are coordinated on one side by three metal ions in the same layer, and on the other side by other halide ions in an adjacent layer (Figure 28.17). This is not what we expect of a solid that is composed of ions. Table 28.7 Structure of some alkaline earth metal halides
198
Figure 28.17 The environment of the halogen in a CdC12 or Cd12 layer structure. Each layer consists of three decks, the central deck containing cadmiums, and the upper and lower decks containing halogens. On one side of each halogen, there are three cadmiums in the same layer at a relatively short distance. On the other side, there are three halogens in the next layer at a considerably larger distance.
In Table 28.7 these troublesome layer structures are printed in green at the top right; you know of two distinct ways of explaining this. One was to argue that the electronegativity difference between the Group I1 element and the halogen is least at the top right of the Table, so it is here that the ionic model is most likely to break down. What is the second type of explanation? At the top right of Table 28.7, small cations such as Mg2+are combined with large anions such as I-. The small cation may then polarize the electron cloud of the large anion, introducing a degree of covalency. Let’s use these explanations to make predictions about beryllium halides, which have not been included in Table 28.7. Will beryllium dihalides conform less closely or more closely to an ionic model than magnesium dihalides? Less closely; using the first argument, beryllium is more electronegative than magnesium, so there will be smaller electronegativity differences in its halides than for magnesium halides. Using the second argument, Be2+is smaller than Mg2+,and will cause more polarization of halide ions. This prediction gives a correct result. Unlike other Group I1 dichlorides and difluorides, molten BeC12 and BeF2 are very poor conductors of electricity. Moreover, the structure of solid BeC12 is even less consistent with an ionic picture than the layer structures of MgC12, MgBr2 and Mg12. It consists of parallel chains in which each beryllium is tetrahedrally coordinated to four chlorines (Figure 28.18). Each chlorine has two beryllium atoms coordinated to it, but in all other directions it is surrounded by chlorine atoms in other chains, which are considerably further away.
Figure 28.18 The chain structure of solid BeC12. I99
BeF2 has a quartz-like structure in which beryllium is again tetrahedrally coordinated. However, the twofold coordination of fluorine is not linear as one might expect in a collection of ions: the LBeFBe bond angle is about 145".
Sodium chloride has a structure in which each sodium is surrounded by six chlorines, and each chlorine is surrounded by six sodiums. Magnesium is sodium's neighbour in the Periodic Table, and its chloride has a CdC12 layer structure. Use the arguments of Section 28.5 to explain this difference.
The Group I1 elements are less willing to form new complexes in solution than the transition elements, but more willing than the alkali metals. Like the alkali metals, for example, their cations complex with polycyclic ethers and cryptands (Structures 27.1,27.2,27.5 and 27.6). Unlike the alkali metals, however, they form complexes in aqueous solution with the hexadentate ligand, edta4- (Table 26.1): M2+(aq) + edta4-(aq) = [M(edta)12-(aq) (28.29) The structure of the calcium-edta complex is shown in Figure 28.19; notice how the edta4- ligand wraps itself around the metal ion so that the latter is octahedrally coordinated by donating atoms with non-bonded electron pairs.
Figure 28.19 The structure of the complex [Ca(edta)]*-.
The Group I1 element that forms complexes most readily with simple unidentate ligands like H20, NH3, OH- and halide ions, is beryllium. One mark of this that you have already encountered is the formation of the complex [Be(0H),l2-, which shows the amphoteric character of Be(OH)2 (Section 28.4). In this and other cases, beryllium shows a marked preference for tetrahedral coordination. Thus, if beryllium metal is dissolved in dilute hydrochloric or sulfuric acid, and if the solutions are evaporated until crystallization of solids begins, the products are the hydrates 200
[Be(H20)4]C12and [Be(H20)4]S0,, respectively. In these compounds, the water of the tetrahedral complex [Be(H20),l2+ is held very tenaciously. Likewise, if BeC12 is recrystallized from liquid ammonia, the product is [Be(NH3)4]C12. Other signs of this preference of beryllium for tetrahedral, fourfold coordination were the structures of BeO, BeC12 and BeF2 (Sections 28.4 and 28.5). It can be linked to the failure of beryllium compounds to match the expectations of an ionic model. Consider, for example, the complex [Be(NH3),I2+.In Section 26.2, we said that if one adopts a covalent picture of complex formation, a metal-ligand bond involves the donation of non-bonded electron pairs on the nitrogen atom of the ammonia ligand to the central metal ion. If so, what type of electronic configuration does the beryllium in [Be(NH3),I2+ have? The configuration of neon, a noble gas: Be2+has the configuration ls2; the four NH3 ligands donate four non-bonded pairs and these eight electrons will occupy the 2s and 2p levels of beryllium to give the configuration ls22s22p6. This explanation implies that beryllium's preference for four-coordination is a mark of covalent bond formation. It therefore links that preference with the structural evidence for the breakdown of an ionic model which we noted in Section 28.4.
1
2
3
4
5 6
Magnesium and calcium are very abundant. They occur mainly as carbonate compounds such as limestone, or as chloride brines from which they can be obtained by evaporation. The metals are made by electrolysis of their molten chlorides. The metals have higher densities, melting temperatures and enthalpies of atomization than the adjacent alkali metals in the Periodic Table. This is because they have two bonding electrons per metal atom rather than one. They are powerful reducing agents that react readily with acids to give hydrogen gas. The reaction with water is negligible or very slow in the cases of beryllium and magnesium because of insoluble, protective oxidehydroxide films. However, calcium, strontium and barium react readily with water, yielding alkaline solutions and hydrogen. In the atoms of Group I1 elements, the nuclear charge is lower than for other elements in the same Period (except for the alkali metals). The first two ionizations, which lead to a noble gas configuration, therefore have relatively low energies. By contrast, the third ionization energies are unusually large because an electron must be removed from an inner shell, whose electrons are closer to the nucleus than the outer s2 electron pair. The combination of relatively low first and second, and high third ionization energies is primarily responsible for the readiness of the Group I1 metals to form substances containing M2+ions, but not ions of higher charge. Group I1 substances containing M+ ions are unknown because they are unstable to disproportionation into the metals and substances containing M2+ions. Limestone, CaC03, yields quicklime, CaO, on heating, and water then converts quicklime into slaked lime, Ca(OH)2.All three compounds can neutralize acids or acidic oxides. 201
Limestone is used in glass manufacture. At high temperatures, carbon dioxide is lost, leaving calcium oxide, which transfers its oxide ion to non-metallic oxides such as SO2, replacing bonds formed with oxygen atoms by bonds formed with oxygens bearing a negative charge. With silica, this generates a silicate anion framework containing cations such as Ca2+and Na+. This melt becomes a glass on cooling. Limestone is slightly soluble in rainwater (which is acidic because of dissolved C02). The consequent erosion creates surface features like cliffs and gorges, and subterranean pores, cracks and fissures in limestone strata, including stalactites and stalagmites. Consequently, these strata both store and transport water. The ionic radii of the M2+ions increase from beryllium to barium. Consequently, the thermal stabilities of the peroxides, hydroxides and carbonates, which decompose to oxides of larger -L value, increase down the Group. Barium, for example, forms a peroxide when the metal is heated in air, but the Group I1 elements above it do not, 10 The Group I1 halides BeX2, MgC12, MgBr2, Mg12 and Ca12 have properties not wholly consistent with an ionic model. For example, BeC12 has a chain structure, and MgC12,MgBr2 and Mg12 have layer structures; molten BeF2 and BeC12 are poor conductors of electricity. These properties can be attributed to polarization brought about by the juxtaposition of small cations and large anions, or to smaller electronegativity differences. 11 The Group I1 ions form complexes with polycyclic ethers and cryptands and with edta4- in aqueous solution. Beryllium is the Group I1 element that complexes most strongly with common unidentate ligands; in doing so, it takes on tetrahedral coordination. In terms of a covalent bonding model, this gives beryllium the electronic configuration of the subsequent noble gas, neon.
Boilers often become lined with ‘scale’, which is deposited from hot water containing dissolved salts. Scale typically consists of sparingly soluble calcium compounds, such as CaC03 and CaS04. One way of removing it is to treat the inside of the boiler with a solution of sodium ethylenediaminetetraacetate. Explain.
Beryllium chloride, BeC12, dissolves in ethylenediamine (Structure 26.2), forming a beryllium complex. If solvent is evaporated, the chloride of this complex eventually crystallizes out. Predict the formula and structure of the complex chloride.
202
The thermodynamics in this Book has been put together for a particular purpose an exploration of the reactions of metals. For this reason, it is less general than other accounts of the subject, and in this Appendix, comments are made on some of the important differences. 1 The chemical reactions that are discussed in this Book take place mainly at constant temperature and constant pressure. Under these conditions, the enthalpy change, AH, and the Gibbs energy change, AG, are the energy changes that are especially informative. They therefore receive unusual emphasis, and the internal energy function, U , is not discussed. 2 At constant pressure, AH is informative because for both a reaction and a phase change of a pure substance, it can be equated to the heat absorbed from the surroundings: AH = q (at constantp) (6.1)
3
In the constant-pressure processes to which Equation 6.1 applies, some work may be done alongside the absorption or evolution of heat. This work arises from volume changes. For example, when water boils at one atmosphere pressure, the volume expands some 1700 fold. The boiling water must create the space for this expansion by pushing back the surrounding atmosphere that presses in on it. This means that the boiling water must do work, akin to the work that must be done in blowing up a balloon. AH is an energy change that is defined so that Equation 6.1 remains correct when this type of work is done. In other words, Equation 6.1 holds provided that the only work done relates to volume change. If other sorts of work are performed as well, then Equation 6.1 must include them. Thus, if electrical work, wel, is performed on the substance under consideration, then Equation 6.1 becomes:
AH = q + w,1
(8.1)
This is our chosen but restricted statement of the first law of thermodynamics. It applies to constant-pressure processes in which the only types of work that are done are electrical work, which appears explicitly in Equation 8.1, and the work of volume change discussed in point 2 . We chose this statement of the first law because it is particularly suited to the discussion of the calorimetric measurements made in Sections 8-10. Equation 8.1 implies that the enthalpy change of a reaction cannot be identified with the heat absorbed if electrical work is performed at the same time. Important examples of this are the self-driving electrical batteries discussed in the Batteries Case Study. These do electrical work for the benefit of the user, and the heat absorbed is then not equal to the enthalpy change of the reaction that takes place inside the battery.
203
4
AG is informative because, at constant temperature and pressure, its sign and size are an indication of how thermodynamically favourable a reaction is. Our decision to emphasize AG in this way gives the equilibrium constant, K, a very low profile in this Book. K can be related to AGE by Equation 11.7 or 11.8; for example
AG:
= -2.303RTlogK
(11.8)
Note, however, that as Table 1 1 . 1 implies, the values of K obtained from AG: values using this equation are dimensionless. You may be used to equilibrium constants with dimensions of concentration (e.g. mol litre-' in Question 4.2); we do not discuss the difference between these two kinds of equilibrium constant here. All that we will say here is that, if the dimensions in the kind of equilibrium constant that you are familiar with are removed, the resulting number is not very different from the number obtained from AG: by applying Equation 11.8.
204
Now that you have completed Metals and Chemical Change, you should be able to do the following things: 1 Recognize valid definitions of and use in a correct context the terms, concepts and principles in the following Table. (All Questions) List of scientific terms, concepts and principles introduced in Metals and Chemical Change
205
2 3
4
5 6
7
8 9 10 I1 12
13 14
15
16
17
Identify instances of oxidation and reduction in terms of loss and gain of electrons. (Questions 2.1 and 2.4) Describe and, if appropriate, write equations for the effect of dilute acid on alkali metals, the alkaline earth metals, zinc, iron, copper, mercury, silver and gold. (Question 2.2) Use the classification of metals made in Sections 2 4 to predict the reaction (or lack of a reaction) when a metal in one class is treated with a halide, oxide or aqueous ion of a metal in a different class. (Questions 2.3 and 4.1) Use criteria derived from the second law of thermodynamics to predict whether a given change should occur or not. (Questions 7.1-7.3) For a change at constant pressure, express the first law of thermodynamics in terms of the enthalpy of the system, and hence determine the enthalpy change for a pure substance. (Question 8. I ) With reference to a constant-pressure calorimeter: (a) relate the experimental measurements made to the accompanying enthalpy changes; (b) relate the measurements in (a) to the enthalpy change at constant temperature for the process under consideration; (c) state any assumption underlying the procedures used. (Questions 8.2 and 8.5) Given the results of a calorimetric experiment, calculate the corresponding molar enthalpy change. (Questions 8.2, 8.3 and 8.5) Show how the molar enthalpy change for a reaction depends on what multiple of the balanced reaction equation is used. (Questions 8.4-8.6 and 12.1) Calculate entropies of fusion or vaporization from enthalpies of fusion or vaporization, and melting or boiling temperatures. (Question 10.1) Use appropriate experimental data to determine entropy changes over a given temperature range. (Questions 10.2 and 10.3) Define the Gibbs function in terms of enthalpy, entropy and temperature, and use this definition to derive an expression relating the standard molar change in the Gibbs function for a reaction, AG:, to the corresponding standard molar enthalpy and entropy changes, AH: and , : S A respectively. (Questions 11.1, 11.3 and 12.1) Write down the formation reaction for a given compound or aqueous ion. (Questions 9.1, 9.2, 11.5 and 12.2) Given a table containing values of AH?, AGyand for elements, compounds and aqueous ions, calculate: (a) standard molar enthalpies of reaction; (b) standard molar entropies of reaction; (c) standard molar Gibbs function changes of reaction. (Questions 9.1-9.4, 10.4, 11.1, 11-3-1 1.5, 12.1-12.3, 13.2, 13.3, 15.1, 15.2 and 17.1) Express the criterion for spontaneous change in terms of standard molar quantities, and hence use the results of the calculations in Learning Outcome 14 to predict whether or not a given reaction is thermodynamically favourable. (Questions 11.1, 11.3, 13.1-13.3, 15.3 and 17.2) Given the appropriate thermodynamic data, analyse statements made about reactivity or lack of reactivity in terms of kinetic and thermodynamic effects. (Questions 13.2, 13.3 and 15.3) Decide when similar reactions have similar entropy changes. (Questions 16.1 and 16.2)
206
18 Given a list of reactions involving gases and solids only, identify those that have: (a) large negative entropy changes; (b) large positive entropy changes; (c) small entropy changes, which may be either positive or negative. (Question 16.2). 19 Use values of AH: and AS: for a reaction at 298.15 K to estimate a value of AGE at higher temperatures. (Question 17.1) 20 Use Ellingham diagrams to predict temperatures at which reactions become thermodynamically favourable. (Question 17.2) 21 Relate the categorization of techniques of metal extraction given in Sections 3 and 17 to the position of metals in the Periodic Table, and to the thermodynamics of metals and their compounds. (Questions 17.3 and 18.4) 22 Construct and use Born-Haber cycles to calculate the lattice energies of metallic halides, (Questions 18.1-1 8.3 and 20.2) 23 Given the formula unit for an ionic compound, use appropriate theoretical expressions to calculate the lattice energy from the appropriate combination of the Madelung constant, ionic charges and ionic radii. (Questions 20.1-20.5) 24 Given a table of standard redox potentials, identify strong and weak oxidizing or reducing agents, and use the table to predict whether redox reactions are thermodynamically favourable or not. (Questions 21.1-2 1.3) 25 Explain and exploit the relationship that exists among the typical elements between electronic configurations, nuclear charges, position in the Periodic Table and ionization energies. (Question 22.1) 26 Recall important items of informhion about the Group I and Group I1 elements, notably those summarized in Sections 23.2,24.3, 25.4, 26.3, 27.2 and 28.7, and combine them to gain new insights about the chemistry of those elements. (Questions 23.1, 25.1, 27.1, 28.1, 28.2, 28.5 and 28.8-28.1 1) 27 Use thermodynamic cycles to relate the relative stabilities and decomposition temperatures of Group I and Group I1 compounds to variations in cation size, and to the lattice energies of solid compounds involved in the decomposition reactions. (Questions 25.2, 25.3, 28.6 and 28.7) 28 Given the crystal structure of a solid containing a complex, identify the complex along with the ligands that it contains, and write equations for reactions in which the coordination of the central element in the complex is maintained, but the ligands are replaced by others. (Questions 26.1, 28.10 and 28.11) 29 Estimate the standard enthalpies of formation of unknown ionic compounds by inserting estimated values of their lattice energies into a Born-Haber cycle, and then search for thermodynamically favourable decomposition reactions that might explain why such compounds do not exist. (Questions 28.3 and 28.4)
207
Reactions (i), (ii), (iii) and (v) are redox reactions. (i) Potassium is oxidized and hydrogen is reduced. The potassium atoms in the metal lose an electron and become K+(aq);the hydrogen ions gain an electron to become hydrogen atoms, which then combine to give H2(g). Copper ions are reduced and metallic iron is oxidized.
(ii)
(iii) MgF2 is an ionic compound formulated as Mg2+(F-)2.Thus, magnesium atoms are oxidized to Mg2+,and fluorine atoms in F2 are reduced to F-. (v) Chlorine atoms in C12 molecules are reduced, and iron in the form Fe2+(aq) loses an electron and is oxidized to Fe3+(aq).
Note that with the ionic formulation Ca2+(F-)2for CaF2, calcium and fluorine retain the ionic forms Ca2+and F- in reaction (iv). This is therefore not a redox reaction.
Rubidium is an alkali metal and strontium is an alkaline earth metal, so both react vigorously with aqueous hydrogen ions and therefore with hydrochloric acid. Hydrogen gas is produced, and Rb+ or Sr2+ions are formed in solution: Rb(s) + H+(aq) = Rb+(aq) + ;H2(g) Sr(s) + 2H+(aq) = Sr2+(aq)+ H2(g)
(Q.1) (Q.2)
Silver was placed in our first class of metals, and so does not react.
Copper and silver are members of our first class of metals, iron and tin are members of the second, and magnesium and calcium are members of the third. A reaction therefore occurs in cases (ii) and (iii): Sn(s) + 2Ag+(aq) = Sn2+(aq)+ 2Ag(s) (Q.3) Mg(s) + Sn2+(aq)= Mg2+(aq) + Sn(s) (Q.4) but not in cases (i) and (iv).
(i)
Tin is oxidized; copper (in the form Cu2+(aq))is reduced.
(ii)
Lead is oxidized; silver is reduced.
(iii) Bromine is oxidized; chlorine is reduced. (iv) Barium is oxidized; chlorine is reduced.
(v) With the formulation Ag+Cl- for AgC1, it is apparent that there is no oxidation or reduction.
208
(vi) Zinc is oxidized; silver is reduced. (vii) Potassium is oxidized; aluminium is reduced. (viii) Aluminium is oxidized; iron is reduced. (ix) Iron is oxidized and oxygen is reduced; using the ionic formulation, you can see that FeO contains Fe2+and 02ions, whereas Fe203contains Fe3+and 02-ions. (x) Iron is oxidized; chlorine is reduced. Notice that in both its compounds and in aqueous solution, iron forms more than one kind of monatomic ion. This is not true of many of the metals that we consider in this Book.
If the qualitative arguments built up in Sections 2-4 are correct, a reaction would be expected in all cases except (i), (iv) and (vi).
+ CuO = ZnO + Cu (iii) Zn + FeC12 = ZnC12 + Fe (v) 3K + AlC13 = 3KC1 + A1 (vii) Zn(s) + 2Ag+(aq) = Zn2+(aq) + 2Ag(s) (viii) Fe + PbBr2 = FeBr2 + Pb (ix) Zn(s) + H20(g) = ZnO(s) + H2(g) (x) Ba + FeBr2 = BaBr2 + Fe (ii) Zn
All parts of this question except (iii), (v) and (viii) can be answered either by reference to the classes in Sections 2.2 and 2.3 or to the series in Section 4. For (iii), (v) and (viii), the two metals are in the same class, and only the series in Section 4 can be used.
The equilibrium constant, K, is given by
Because acetic acid is a weak acid, equilibrium lies to the left, and a solution of acetic acid contains only small concentrations of hydrogen and acetate ions; HAc(aq) is the predominant species in solution. This is marked by the fact that K is small; the actual value is 1.8 x mol litre-*.
The value of AH quoted in the question is negative, so the reaction is exothermic. According to the principle of the conservation of energy, the reverse reaction, the decomposition of one mole of NaCl, must absorb an equivalent amount of energy. Hence it must be endothermic; that is NaCl(s) = Na(s) + iC12(g);AH = +411.2kJ (Q.6) According to these values, solid sodium chloride is of lower enthalpy than a mixture of sodium metal and chlorine gas (at constant temperature and pressure), so Thomsen’s hypothesis could explain the behaviour of this system.
209
This time, block 2 absorbs heat q, so AS2 = q/T2,and block 1 loses heat q, so AS1 = -q/T1. Thus
Taking the two blocks to be an isolated system, AStotalis negative because T2 > T I . This violates the second law: the process cannot occur.
For Reaction 5.2 Na(s) + +C12(g) = NaCl(s) AS = -90.7 J K-l and AH = -41 1.2 kJ, so
Thus
AH
AS - -= (-90.7 - (- 1379))J K-' T = + I 288 JK-'
The value is positive, so the second law is vindicated!
(i) In this case the reaction is endothermic, so heat is withdrawn from the surroundings; AS,,, must be negative: AS,,, =-- AH - - 471005 T 298.15K
(ii)
AStotal
= AS,,, + A&-€!ac. = (-158.0 + 334.6) J K-' = +176.6 J K-'
The total entropy change is positive, in accord with the requirements of the second law. From the inequality 7.10, the quantity (AH- TAS) must be negative for a reaction to take place (at constant T and p ) . AH is positive for an endothermic reaction, so the term TAS must be both positive and large enough to outweigh this contribution. Since T is positive by definition, AS must also be positive, as in the example above.
The electrical work done on the system is given by wel = power x time = (100 W) x (60 + 44) s = (lOOJS-') x (104s) = 10.4 x 1035 = 10.4kJ 210
For a change at constant pressure AH = q + wel Assuming that the system is perfectly insulated, q = 0. So AH = wel = 10.4kJ Notice that this value of AH actually refers to the system as whole, including the container, heating coil and thermometer, as well as the water. In any accurate experiment to determine the enthalpy change for the water alone, allowance would have to be made for the other components.
(a) For a change at constant pressure AH = q + w,1 Assuming perfect insulation, q = 0, so AH = wel. The electrical work done is given by w,1 = (50 J S-') x (81 S) Thus, for vaporization of 1.62g of water, AH=4050J (b) AH for a process like this depends on the amount of water vaporized. One mole of H 2 0 has a mass of (2 x 1.007 9 + 15.999 4) g or 18.02 g. Thus, 1.62 g amounts to (1.62 g/l8.02 g) mol or 0.089 9 mol. Then AH for 1 mol is given by
AH=
4050 J 0.0899 mol
= 45.1 x lo3J mol-I = 45.1 kJ mol-'
As you will see in Section 8.3, this value of AH is generally called the molar enthalpy of vaporization, denoted by AHm.The generally accepted value for water is AHm= 44.3 kJ mol-l: the value obtained above is too high because we neglected heat loss from the system.
To calculate the molar enthalpy change for the reaction
H+(aq) + OH-(aq) = H20(1) you must assume that all of the acid and base reacted (they were present initially in equal amounts). Now, 0.1 litre of 2.0 mol litre-' acid were used, which contained 2.0 mol litre-' x 0.1 litre = 0.2 mol of H+(aq)
If all the acid is consumed, then the 'amount of reaction' is 0.2 mol. From the text, AH = -1l.Ok.J. so
AH m
=-&--I
kJ - 55.0 kJrnol-I
0.2 mol
Notice that exactly the same result would be obtained by considering the reaction of all the base (OH-, aq) initially present.
21 I
Since AH = -149.5 kJ when 0.50 mol of lithium reacts completely with oxygen to give 0.25 mol of Li20, we can write: 1
zLi(s)
1 + ,02(g)
= ILi20(s); 4 AH, = -149.5 kJmol-1
(Q.8)
where AHmin this case refers to the balanced equation containing half a mole of lithium. The values of AHmrequired can then be obtained by multiplying this balanced equation, and its AHmvalue, by four, eight and sixteen, respectively: = Li20(s); AHm = -598.0kJmol-1
= 2Li20(s);AHm = -1 196.0 kJ mol-' = 4Li20(s);AHm = -2 392.0 kJ mol-*
This question is similar to the example discussed in the text: again there are two parts to the experiment, both conducted at constant pressure in an insulated container. When the reaction takes place under these conditions, the corresponding enthalpy change is given by AH1 = H(products, 299.5 K) - H(reactants, 298.15 K) In this part of the experiment, no electrical work is done (wl= 0) and, assuming perfect insulation, no heat is transferred (ql = 0). So AH1 = 41 + w1
=o The calibration experiment corresponds to the enthalpy change AH2 = H(products, 299.5 K) - H(products, 298.15 K) In this part of the experiment, electrical work is done on the system, so W2
= We1 = (50 J S-') x (61.5 S)
= 3075J
Again, the calorimeter must be assumed to be perfectly insulated, so q2 = 0 and A H 2 = 42 + w2 = 3075J
Then the enthalpy change for the reaction at 298.15 K is given by AH = H(products, 298.15 K) - H(reactants, 298.15 K) = AH,- A H 2 = -3 075 J
212
(i) One mole of zinc has a mass of 65.38 g, so assuming all of the zinc is consumed, the amount of reaction is (1.308/65.38) mol. For the reaction Zn(s) + 2H+(aq) = Zn2+(aq)+ H2(g)
AH m
= -
3 075 J (1.308/65.38) mol
3 075 x 65.38 Jmol-l 1.308 = -153.7 x 103Jmo1-I = -153.7 kJ mol-'
- -
(ii) The equation -Zn(s) 1 + H+(aq) = iZn2+(aq)+ kH2(g) 2 is the result of dividing the equation in part (i) of this question by two. The value of AHmmust therefore also be halved, so AHm = -76.9 kJ mol-l.
The reaction implied by the question could be Al(s) + SCl,(g) = AlC13(~) or 2Al(s) + 3C12(g) = 2AlCl,(s) or any multiple or sub-multiple of these expressions. Unless an equation is specified, a molar enthalpy change cannot be calculated. The question gives no indication of the equation referred to.
From Table 9.1, the Mf:values are -1 675.7 kJ mol-1 and -1 206.9 kJ mol-1, respectively. They are the AH: values of the reactions in which one mole of A1203(s) and CaC03(s) are formed from their elements in the standard reference states: 2Al(s) + O&) = A1203(s); AH: = -1 675.7 kJ mol-1 (Q.14)
4
Ca(s)
+ C(graphite) +
:02(g) = CaC03(s); AH:
= -1 206.9 kJmol-1
(Q.15)
Notice that graphite rather than diamond is indicated in the CaC03 equation. This is because graphite and not diamond is the standard reference state of carbon, a fact apparent from its zero AH? value in Table 9.1.
(i) For the reaction Na(s) + iC12(g) = NaCl(s) the Data Book tells us that My(NaC1, s, 298.15K) = -411.2kJmol-1 and as Equation 5.2 represents the reaction in which sodium chloride is formed from its elements, the required value of AH: is simply -41 1.2 kJ mol-1 (cf. the value given in Question 6.1). As Na(s) and C12(g) are the reference states for these two elements, their standard enthalpies of formation are zero by definition. 213
(ii) For the reaction PbO(s) + C(graphite) = Pb(s)
+ CO(g)
from the discussion in the text, summarized by Equation 9.18, we have AH: = AH?(Pb, s) + AHF(C0, g) - AHy(Pb0, s) - AH?(C, graphite) Again, the enthalpies of formation of the elements in their reference states, Pb(s) and C(graphite), are zero by definition. Using values from the Data Book, AH: = ( 0 + (-110.5) - (-219.0) - 0 ) kJmol-l = +108.5 kJ mol-1 (iii) Looking at the values for the equation SOC12(1) + H20(1) = S02(g) + 2HCl(g)
(6.4)
we find that none of the entries is zero, and that AH? for HC1 must be multiplied by two. Hence AH: = AHF(S02, g) + 2AH?(HC1, g) - AH7(SOC12,1) - AHF(H20,l) = ((-296.8)
+ 2 x (-92.3)
-
(-242.7) - (-285.8)) ldmol-l
= +47.1 kJ mol-1 (cf. the value given in Section 6)
Notice that a different value would be obtained if gaseous water were specified. (iv) For the reaction 3K(s) + AlC13(s) = 3KCl(s)
+ Al(s)
(3.9)
the entries for K(s) and Al(s) are zero, so AH: = 3 AHY(KC1, S) - AHF(AlCl,, S) = { 3 x (-436.7) - (-704.2)) kJmol-1 = -605.9 kJ mol-1 Notice that the values in parts (i) and (iii) are those quoted in Section 6: they were standard molar enthalpy changes.
(i) For the equation Zn(s) + 2H+(aq) = Zn2+(aq)+ H2(g) all the values of AH? are zero by definition except that of Zn2+(aq), so
AH:
= AHy(Zn2+, aq) = -153.9 kJ mol-1
The important implications of this result are taken up in Section 11.2. (ii) For the equation Mg(s) + 2Ag+(aq) = Mg2+(aq) + 2Ag(s) the entries in the AH? column are zero for the elements, Mg(s) and Ag(s), so AH: = AH? (Mg2+,aq) - 2 AH? (Ag+, aq) = { (-466.9) - 2 x (105.6)) kJmol-l = -678.1 kJ mol-1
214
(2.13)
(iii) For the equation H+(aq) + OH-(aq) = H20(1) the entry for H+(aq) is zero by definition. Thus
AH:
= AHY(H20,l) - AHF(OH-, aq) = { (-285.8) - (-230.0)) kJ mol-1 = -55.8
kJ mol-1 (cf. the value calculated in Question 8.3)
The reactions in the question are: Mg(s) + C12(g) = MgC12(s); AH: = -641.3 kJmol-1 MgC12(s) = Mg2+(aq)+ 2Cl-(aq); AH: = -160.0 kJ mol-1 So, for the sum of these two reactions, Mg(s) + C12(g) = Mg2+(aq) + 2Cl-(aq) AH: = (-641.3 + (-160.0)) kJmol-I = -801.3 kJ mol-l
(9.27) (Q. 16)
(Q.17)
Now, AH?(Cl-, aq, 298.15 K) = -167.2 kJmol-l. So AHF(Mg2+,aq, 298.15 K) = (-801.3 - 2 x (-167.2)} kJmol-l = -466.9 kJ mol-l This is the entry against Mg2+(aq)in the Data Book.
(a) From a generalization of Equation 10.3:
AH,,,= 6.4 x lo3Jmol-'
AS,,, =-
Tfus
172K
= 37.2 J K-1 mol-l
(b) Again from Equation 10.3: AS vap
m v a p
=--
Tvap
- 20.4 x lo3Jmol-'
239 K
= 85.4 J K-1 mol-1
Notice that the positive values of ASmfor fusion, vaporization (and sublimation, solid + gas) are an inevitable consequence of the fact that these processes are endothermic. (c) The answer to part (b) is the molar entropy of vaporization. One mole of Cl2 has a mass of 70.9 g, so 7.09 g represents 0.100 mol. Thus, AS = (85.4 J K-l mol-1) x (0.100mol) = 8.54JK-'
215
(a) The increment covers the temperature range from 20.64 K to 25.22 K, for which the mean temperature, T', is 22.93 K. Then,
(b) Equation 10.5 now tells us that we can obtain an approximate value of the entropy difference between one mole of chlorine at 90.00 K and at 13.50 K by adding up the values of AS' in column 6 of Table 10.2. The result is 39.388 J K-l.
The entropy change is the area beneath the curve between 0 K and 90 K, and this is the sum of areas a, b, c, d and e. area of rectangle u = (90 - 14) K x (0.24 J K-*) = 76 x 0.24JK-l = 18.24JK-' area of rectangle b = (90 - 23) K x (0.45 - 0.24) J K-2 = 67 x 0.21 J K-' = 14.07JK-1 area of triangle c = base x height = 0.5 x (14 K) x (0.24 J KP2) = 1.68 J K-'
area of triangle d
= 0.5 x (23 - 14) K x (0.45 - 0.24) J K-2 = 0.5 x 9 x 0.21 J K-1 = 0.95 J K-'
estimated area of 'triangle' e = 0.5 x (90 - 23) K x (0.61 = 0.5 x 67 x 0.16 J K-1
- 0.45) J K-2
= 5.36JK-l SO AS = (18.24 + 14.07 + 1.68 + 0.95 + 5.36) JK-1 = 40.3 J K-1
The results in Figure 10.4 refer to 1 mol of chlorine, so the molar entropy increment is ASm = 40.3 J K-1 mol-1 . Clearly, this method is approximate: the area e, for example, deviates considerably from a triangle. A more accurate but tedious approach is to count the little squares beneath the curve, including those that are more than half below, and rejecting those that are more than half above. This gives a value of about 40.8 J K-1 mol-1.
(a) (i) Na(s)
+
iC12(g) = NaCl(s)
By Equation 10.I 1, and taking values from the Data Book: AS: = P(NaC1, s) - F ( N a , s) - i p ( C l , , g) = (72.1 - (51.2) - 6(223.1)}JK-lmol-1 = -90,7 J K-1 mol-1 (cf. the value quoted in Section 7.3) Notice two points. Firstly, the values for elements are not zero: absolute entropies are listed in the Data Book, not entropies of formation. Secondly, 216
(5.2)
the subscript m on AS: now refers specifically to the equation as written in the question. If the equation were written: 2Na(s) + C12(g) = 2NaCl(s) (Q. 18) then
= -181.4 JK-l mol-1
AS:
Ifyou got this question right and you do not have time to try all the remaining examples, do one from part ( b )and then look at the values of AS: below, before returning to the main text.
(ii) Mg(s) + 102(g) = MgO(s) AS: = P ( M g 0 , S) - P ( M g , S) - + p ( 0 2 , g) = (26.9 - 32.7 - :(205.1)} J K-' mol-1 = -1 08.4 J K-' rno1-l (iii) CaCO,(s) = CaO(s) + C02(g) AS: = P ( c a 0 , s) + Se(co2, g) - P(CaC03, s) = { 39.7 + 213.7 - (92.9)) J K-l mol-1 = 160.5 J K-1 mol-I (iv) N204(g) = 2N02(g> AS:
(2.5)
(9.10)
(9.19)
= 2p(NO2, g) - p ( N 2 0 4 , g)
= { 2 x (240.1) - 304.3) J K-l mol-1 = 175.9 J K-1 mol-'
(b) (v) Zn(s) + 2H+(aq) = Zn2+(aq) + H2(g) AS: = p ( Z n 2 + ,aq) + p ( H 2 , g) - p ( Z n , s) - 2 p ( H + , aq) = (-112.1 + 130.7 - 41.6 - 0)JK-lmol-1 = -23.0 J K-1 mol-1 (vi) Cu(s) + 2H+(aq) = Cu2+(aq) + H2(g) AS: = p ( C u 2 + ,aq) + p ( H 2 , g) - ~ ( C Us), - 2 p ( H + , aq) = (-99.6 + 130.7 - 33.2 - 0) J K-1 mol-1 = -2.1 J K-l mol-1
(a) SOCl,(l) + H20(1) = S02(g)+ 2HCl(g) AS: = P ( S 0 2 , g) + 2P(HC1, g) - P(SOC12,l) - Se(H20,l) = (248.2 + (2 x 186.9) - 217.5 - 69.9) J K-1 mol-1 = 334.6 J K-1 mol-1 (b) AG:
=
AH:
-
TAS:
(2.9)
(6.4)
(1 1.4)
= 47.1 kJ mol-1 - 298.15 K x (334.6 J K-l mol-I) = 47.1 kJ mol-1 - 99 761 J mol-1 = (47.1 - 99.8) kJ mol-1 = -52.7kJ mol-1
(c) For Reaction 6.4, AH: is positive so, as noted on p. 44, the occurrence of the reaction violates Thomsen's hypothesis. However, because AS: is large and positive, our calculated value of AG: turns out to be negative. Thus, the essential requirement for a thermodynamically favourable reaction, AGE c 0 is fulfilled. 217
From Equation 11.7, Equation 11.8 or Table 1 1.1, if AG: = 0, then K = 1. The equilibrium constant for the reaction A = B has the form [BI K = IAI If K = I , then the concentrations of A and B at equilibrium must be the same. Since B is formed from A, this must mean that a half or 50% of the reactant remains. This question highlights an extremely important point. AG: = 0 does not mean that the reaction cannot happen: it simply does not go to completion.
(i) Na(s)
+
;C12(g) = NaCl(s)
AH:
= -41 1.2kJ mol-I
AS:
= -90.7 J K-' mol-I
AG:
= -41 1.2 kJ mol-I - (298.15 K) x (-90.7 J K-' mol-l)
(Question 9.2) (Question 10.4)
= (-4 11.2 + 27.0) kJ mol-I = -384.2
kJ mol-'
According to Table 11.1, this large negative value suggests that the reaction should go effectively to completion under ambient conditions. Conversely, the reverse reaction should be undetectable in any practical sense. These results confirm the qualitative insights developed in Section 5.
+ 2H+(aq) = Cu2+(aq) + H2(g)
(ii) Cu(s)
All values of AH:
are zero by definition, save that for Cu2+(aq),so,
AH:
= AHy(Cu2+, aq) = 64.8 kJ mol-I
AS:
= -2.1 J K-' mol-1 (Question 10.4)
AG:
= 64.8 kJ mol-1 = (64.8
(2.9)
-
(298.15 K x -2.1 J K-1 mol-I)
+ 0.63)I~Jmol-~
= +65.4 kJ mol-I
Now you can interpret the lack of reaction when copper is added to acid, in a precise way. Evidently the reaction is unfavourable from a thermodynamic point of view, so you would not expect anything to happen. (iii) Zn(s)
AH: AS:
+ 2H+(aq) = Zn2+(aq) + HZ(g)
(2.4)
= -153.9 kJ mol-1 (Question 9.3) = -23.0 J K-1 mol-1 (Question 10.4)
A G E = -153.9 kJ mol-1 - (298.15 K x -23.0 J K-1 mol-1) = (-153.9
+ 6.86) kJmol-1
= -147.0 kJ mol-I
The value is large and negative. The reaction happens in practice (as you saw in Activity 2.1) because it is both thermodynamically favourable and has a reasonable rate of reaction. This point is examined more fully in Section 13. 218
Using Equation 11.12: AG: = AGF(S02, g)
+ 2AG?(HC1,
+ 2 x (-95.3)
= (-300.2
-
g)
-
AGF(SOCl2,l) - AGF(H20,l)
(-201.0) - (-237.1)) kJmol-1
= -52.7 kJ mol-1
This is exactly the value obtained from AH:
and AS:
in Question 11.1.
For the aqueous ion Sc3+(aq),the formation reaction is Sc(s) + 3H+(aq) = Sc3+(aq)+ iH2(g) so AS? = s8(Sc3+,aq) + :F(H,, g) - ~ ( S Cs), - 3s8(H+, aq) = { -255.2
+
(Q. 19)
(130.7) - 34.6 - 0}J K-I mol-I
= -93.8 J K-I mol-1
A G =~ AH?
-
m? =
+ TAS?
(11.9)
TAS:
so
AG?
= (-586.6 kJ mol-l)
+ (298.15 K) x (-93.8
J K-' mol-l)
= (-586.6 - 28.0) kJ mol-1 = -614.6 kJ mol-1
For the oxide, the appropriate formation reaction is 2Sc(s) + iO,(g) = SC203(S)
so AS? = s8(sc203,sj - ~ P ( s c S) , -
+Se(o,,g)
= { (+77.0) - 2(34.6) - i(205.1)) J K-1 mol-1 = -299.9 J K-1 mol-1
Then AG? =
AH?
-
TAS~
= (-1 908.8 kJ mol-1) - (298.15 K) x (-299.9 J K-' mol-l)
= (-1 908.8
+ 89.4) kJ mol-l
= -1 8 19.4 kJ mol-1
(i) -150.6 kJ mol-1; (ii) 520.4 kJ mol-l; (iii) -233.2 kJ mol-I; (iv) 1 463.4 kJ mol-l. (i) Subtract the silver equation and its AG: A G value. ~
value from the zinc equation and its
(ii) Subtract the magnesium line from the copper line, and then multiply by two. (iii) Subtract the silver line from the iron line, and then multiply by two. (iv) Subtract the aluminium line from the mercury line, and then multiply by six. 219
The completed entries are in Table Q. 1.
Table Q.l Thermodynamic data at 298.15 K for La3+(aq)and Pu3+(aq)
(i) For the two ions, the AH
7 values are the values of AH:
for the reactions
La(s) + 3H+(aq) = La3+(aq) + iH2(g)
Pu(s)
+ 3H+(aq) = Pu3+(aq)+
(12.5)
(12.6)
gH2(g)
This can be shown for the lanthanum reaction in the following way: AH: = ZAHy(products) - CAHy(reactants)
As AH:
= AH? (La3+,aq)
+ 5 AH?
= M?(La’+, aq)
+o-o-o
(9.18)
(H2, g) - AH? (La, s) - 3 AH? (H+, aq)
= -709.0kJmol-1
AH7(La3+, aq) = -709.0 kJ mol-1 Similarly, AH?(Pu3+, aq) = -593.0 kJ mol-1 (ii) Likewise, the AGy values for the two ions are the values of AGE for the reactions (12.5) La(s) + 3H+(aq) = La3+(aq) + ;H2(g) (1 2.6) Pu(s) + 3H+(aq) = Pu3+(aq) + i-H2(g) This can be shown for the lanthanum reaction in the following way: AG: = Z AG? (products) - C AGy(reactants) = AGF(La3+, aq)
+
AGF(H2, g ) = AGF(La3+, aq) + 0 - 0 - 0
For the lanthanum reaction AH: = -709.0 kJ mol-1 and AS: A G =~ AH: -
-
AGy(La, s) - 3 AGy(H+, aq)
= -84.1 J K-1 mol-’
TASE
= -709.0 kJ mol-1 - (298.15 x -84.1) J mol-1
= -709.0
+ 25.1 kJmol-1
= -683.9 kJ mol-1
For the plutonium reaction AH: = -593.0 kJ mol-1 and AS: = -25.9 J K-1 mol-I AGE = -593.0 kJ mol-1 - (298.15 x -25.9) J mol-1 = -593.0 + 7.7 kJ mol-l = -585.3 kJ mol-1
220
(11.12)
(iii) For the lanthanum reaction AS: = -84.1 J K-I mol-I
+ i p ( H 2 , g) - p ( L a , s) - 3 p ( H + , aq) = p(La3+, aq) + (196.1 - 56.5 - 0) J K-1 mol-1 p(La3+,aq) = (-84.1 - 196.1 + 56.5) J K-1 mol-l = p(La3+,aq)
= -223.7 J K-I mol-1
For the plutonium reaction p(Pu3+, aq) = (-25.9 - 196.1 + 50.3) J K-1 mol-1 = -171.7 J K-I mol-1
According to the calculations in Question 12.2: G : La(s) + 3H+(aq) = La3+(aq) + ;H2(g); A
= -683.9 kJ mol-I
(12.5)
Pu(s) + 3H+(aq) = Pu3+(aq) + ;H2(g); A G :
= -585.3 kJ mol-1
(12.6)
Converting the equations into a form in which there is one mole of H+(aq) on the left and half a mole of H2 on the right: fLa(s) + H+(aq) = iLa3+(aq)+ i-H2(g); A G: = -228.0 kJ mol-1 (Q.2 1) iPu(s) 3 + H+(aq) = fPu3+(aq)+ iH2(g); AG:
= -195.1 kJ mol-1
(Q.22)
These values put lanthanum between sodium and magnesium, and plutonium between magnesium and aluminium.
Reactions (i) and (iii) have negative values of AGE,so in these cases, equilibrium lies to the right. Ignoring the possibility of slow rates of reaction, these reactions should occur.
From Table 12.1, for the first reaction AG: = 2{32.7 - (-227.5)) kJ mol-I = 520.4 kJ mol-I
and for the second reaction A G : = 6(-161.7 - 32.7) kJ mol-1 = -1 166.4 kJ mol-'
Thus, in the first case, the lack of reaction is due to thermodynamic stability; in the second case it must be due to kinetic stability. In the second case we can attribute the kinetic stability to the oxide coating on the aluminium.
22I
(a) A conceivable reaction here would be C(diamond) + 02(g) = C02(g) From the Data Book we find AGE = AGF(CO2, g) - AGF(C, diamond) - AGF(02, g) = (-394.4
-
2.9 - 0) kJ mol-1 = -397.3 kJ mol-1
(b) A conceivable reaction would be Mg(s) + ZnC12(s) = MgC12(s) + Zn(s) From the Data Book we find AGE = AGF(MgC12, s) + AG?(Zn, s) - AGF(Mg, s) - AGY(ZnCl2, s) = (-591.8 + 0 - 0 + 369.4) kJ mol-1 = -222.4 kJ mol-1 Thus, in both cases, there is a thermodynamically favourable reaction that the system could undergo: both systems are therefore kinetically stable.
(i) Subtract the lead reaction from the aluminium reaction in Table 15.1, and then multiply by three: AGE = -1 015.5 kJmol-1 (ii) Subtract the lithium line from the copper line: AG:
= 43 1.5 kJ mol-1
The values of AGF for the chlorides must be converted into values corresponding to the reaction of the different metals with the same number of moles of chlorine, C12(g). In Table Q.2 the values of AGE corresponding to half a mole of chlorine have been used. Thus, for sodium, AG: is equal to AGF(NaC1, s), but for aluminium, AG: is equal to AGy(AlCl3, s). Table Q.2 Values of AGE for the reactions of metals with chlorine at 298.15 K
222
Table Q.2 does support the conclusions reached in Section 15. There are differences from the orders in Tables 12.1 and 15.1 in fine detail, but there are also broad similarities. Thus, sodium and calcium come near the top, but copper, silver and mercury are near the bottom.
Reactions obtained by subtracting equations in Table 15.1 involve the reaction between solid metals and solid oxides. Reactions between solids tend to be slow because they can occur only where the surfaces of the different solids make contact. By contrast, reactions obtained by subtracting equations in Table 12,l involve the reaction between solid metals and aqueous ions. Fresh ions are constantly being brought up to the metal surface by the turmoil in the solution, so these reactions tend to be faster than solid-solid reactions.
The pair of reactions with similar values of AS: is the pair that are analogous in the sense defined at the end of Section 16. Pair (i) does not fit this description because, although the reactions differ in the substitution of mercury for copper and the numbers in the equations are identical, the elemental mercury is gaseous and the copper is solid. Pair (iii) is not the answer because the numbers preceding 02(g) are not identical. Pair (ii) is the correct answer: it contains two reactions with the same physical states and coefficients; they differ only in the substitution of magnesium by calcium.
(a) Reaction (i); (b) reactions (iv) and (v); (c) reactions (ii) and (iii). Reaction (i) involves a decrease in the number of moles of gas. Reactions (ii) and (iii) involve no change in the number of moles of gas. Reactions (iv) and (v) involve an increase in the number of moles of gas.
Note that mercury is in the gaseous state. For the decomposition reaction at 298.15 1K:
AH:
1
= @(Hg, g> + m302,g> - M Y ( H g 0 , s) = (61.3 + 0) - (-90.8) kJ mol-1 = 152.1 kJ mol-1
AS:
+
$Se(O,, g)
Se(Hg0, S) = (175.0 + 102.6 - 70.3) J K-1 mol-1 = 207.3 J K-' mol-1 = Se(Hg, g)
-
Now use Equation 17.3: T=
M:(298.15K) AS: (298.15 K)
152.1kJ mol-* = 734K 207.3 J K-' mol-I This temperature is close to the value of about 750 K, at which the value of AG: for the HgO line on Figure 17.5 is zero (above the boiling temperature of mercury). -
223
In Figure 17.5, the CO line drops below the MnO line at about 1 700 K, and below the ZnO line at about 1 200 K. It is above these temperatures that carbon is thermodynamically capable of reducing the oxides. At such high temperatures we expect the reactions to be fairly fast, and reduction does in fact occur in both cases. In the zinc reaction the crossover point of the ZnO and CO plots occurs above the boiling temperature of zinc. The product is thus zinc vapour and the reaction is
C(s)
+ ZnO(s) = CO(g) + Zn(g)
(Q.23)
For Mn at 1 700 K the reaction is similar except that the product would be liquid manganese.
Statements (i), (iii) and (v) are false; statements (ii), (iv) and (vi) are true. (i) Those metals whose oxides have the most negative values of AGF are hardest to extract from oxides, so from Table 15.1, copper is easier to extract than tin. (ii) See Section 17.2. (iii) Only those metals whose compounds have the least negative values of AG? are found free in nature. The lanthanides and actinides are not in this category, as, in fact, Question 12.3 and Figure 17.7 suggest. (iv) Calcium is near the top in Tables 15.1 and Q.2, so it should be thermodynamically capable of reducing the compounds of many other metals. (v) Only those metals whose oxides or chlorides have the least negative values of A($ are obtained by heating their oxides or chlorides. (vi) Electrolysis of fused salts is expensive. It is used only where compounds of the metals have such negative values of AGF that normally cheaper methods such as carbon reduction become expensive because very high temperatures are needed.
Step a, (vi); step b, (ix); step c, (iv); step d, (vii); step e, (x); step f, (ii).
From Figure 18.2, AHy(LiF, s) = AH2m(Li, s) + II(Li) + :D(F-F) -616kJmol-1 = (159 + 520 + 79 - 328) kJmol-1 L(LiF, s) = -1 046 kJ mol-1
E(F) + L(LiF, s) + L(LiF, s) -
This value is considerably more negative than that for NaCl; thus more energy is released when lithium fluoride is formed from its ions than when sodium chloride is formed from its ions. The reason for this is discussed in Section 20.4.
224
Figure Q.l The Born-Haber cycle for calcium chloride, CaC12.
In the Born-Haber cycle for CaC12, which is shown in Figure Q. 1, the lattice energy corresponds to the reaction Ca2+(g) + 2C1-(g) = CaC12(s) The process Ca(g) = Ca2+(g) + 2e-(g) is the sum of the processes Ca(g> = Ca+(g>+ e-(g> Ca+(g) = Ca2+(g) + e-(g) whose ionization energies are ZI(Ca) and Z2(Ca), respectively. The two electrons are taken up by the two chlorine atoms. From the cycle my(CaC12, s) = M z m ( C a , s) + II(Ca> + 1 2 ( ~ a )+ D(Cl-cl) 2E(C1) + L(CaC12, s) -796kJ mol-I = (178 + 590 + 1 145 + 244
-
698) kJmol-1
-
+ L(CaC12, s)
Therefore L(CaC12, s) = -2 255 kJ mol-I Thus, about three times as much energy is released when one mole of solid calcium chloride is formed from gaseous Ca2+and C1- ions as when one mole of solid sodium chloride is formed from gaseous Na+ and C1- ions. If we assume that the bonding in both chlorides is ionic, then the attraction between the ions is larger in CaC12 because the solid contains doubly charged Ca2+ ions. Another reason for the difference is that one mole of CaC12 contains 50% more ions than one mole of NaC1, so the energy of interaction will be correspondingly greater. These ideas are discussed quantitatively in Section 20.4.
Those metals with low ionization energies tend to form compounds with the most negative values of AGF. These metals lie near the top of series such as those in Tables 12.1 and 15.1, and are not easily extracted from their compounds. Such metals lie to the left of Figure 17.7, so one expects ionization energies to be low at the left of the Periodic Table and high at the right. This is broadly speaking correct, as a comparison of, say, rubidium and silver shows.
225
For LiF, with n = 6, L(LiF, s ) = -
--
1.389 x lo5 x 1.748 X 1 x 1 201 1 . 3 8 9 lo5 ~ x 1.748 x 1 x 1 x 5 kJ mol-' 201 x 6
= -1 007 kJ mol-I
For CaC12, with n = 9, 1.389 x lo5 x 2.408 x 2 x 1 L(CaC12, s) = (1- !-]kJ 274 = -2 170kJ mol-1
mol-'
The results are the same as those in the penultimate column of Table 20.2.
The cycle is shown in Figure Q.2. It corresponds closely with that given for NaI in Figure 20.8. However, the standard state of bromine at room temperature is Br2(1), so iBr2(1) replaces ; I ~ ( S )on the left-hand side of the formation reaction for the solid halide. And K, of course, replaces Na.
Figure Q.2 Born-Haber cycle for potassium bromide. KBr.
From the cycle, AHF(KBr, s) = AH&(K, s)
+ ZI(K) +
AHy(Br, g) - E(Br) + L(KBr, s)
Substituting values from the Data Book: -393.8kJmol-1 = (89.2 + 419 + 111.9 - 324)kJmol-1 L(KBr, s) = -690 kJ mol-1
+ L(KBr, s)
Note the use of A@(Br, g), which is greater than $D(Br-Br). This is because Br2 is a liquid at 25 "C (see Section 20.5). K+ has the configuration of argon (n = 9) and Br- has the configuration of krypton (n = 10). Thus, for KBr we use the average, n = 9.5:
226
Questions: answers and comments
L = - 1.389 x lo5 x 1.748 x 1 x 1 330 - - 1.389 x
1
1O5 x 1.748 x 0.895 kJ mol-' 330
= -658 kJ mol-I
As usual, the value is not as negative as the cycle value, but it deviates from it by only 32 kJ mol-I, which is less than 5%.
Ca2+has the configuration of argon ( n = 9), and the two F- ions in the formula unit each have the configuration of neon ( n = 7). The average value for the three ions is i ( 9 + 7 + 7) = 7.67. Table 20.1 tells us that for fluorite, M = 2.5 19 so:
L = - 1.389 x lo5 x 2.519 x 2 x 1 [l - &)kJnlull 237 - - 1.389 x lo5 x 2.519 x 2 x 0.870 kJ mol-l 237 = -2 569 kJ mol-1 This value is not negative enough, but the discrepancy is only 66 kJ mol-I, which is less than 3%.
In the formula unit CaF2, there are three ions so v = 3. Hence 1.079 x 105 x 3 x 2 x I kJ mol-l L(CaF2, s) = (114+119) = -2 779 kJ mol-'
This value is 144 kJ mol-1 too negative. The agreement is worse than with the BornLand6 equation, the discrepancy being about 5.5%.
E?(A13+ I Al) = -1.68 V. First reverse the equation, and the sign of the AGE value: A13+(aq)+ ;H2(g) = Al(s) + 3H+(aq); AG: = 485 kJ mol-I (Q.24)
Use the symbol e as shorthand for iH2(g) on the side of the equation on which it appears, and for H+(aq) on the other side. Since there is H2(g) on the left, and 3H+(aq)on the right, the equation now becomes: A13+(aq)+ 3e = Al(s); AGE = 485 kJ mol-1 (21.13) Using Equation 2 1.4: 485 kJ mol-I =-1.68V 3 x 96.485 kJ mol-' V-'
p=-
This value is considerably more negative than the threshold value of -0.1 V, so we conclude that aluminium is a powerful reducing agent.
227
The two strongest oxidizing agents are fluorine, F2, and chlorine, C12, and the two weakest are the aqueous ions Li+(aq) and Cs+(aq). The two strongest reducing agents are lithium metal and caesium metal, and the two weakest reducing agents are the aqueous fluoride and chloride ions. The reducing agents occur on the right of the equations in Table 21.2 and the oxidizing agents on the left. The most powerful reducing agents (and therefore the least powerful oxidizing agents) are found in systems with the most negative E? values. By contrast, the least powerful reducing agents (and therefore the most powerful oxidizing agents) are found in systems with the most positive @ values.
Reactions (i), (ii), (iv) and (v) are thermodynamically favourable. The equations are: (i) Mg(s)
+ Zn2+(aq)= Mg2+(aq)+ Zn(s)
+ Sn2+(aq)= Zn2+(aq)+ Sn(s) (iv) Fe2+(aq)+ +C12(g) = Fe3+(aq) + C1-(aq) (v) W a q ) + kC12(g) = tBr2(aq) + C1-(aq) (ii) Zn(s)
(Q.25) (Q-26) (Q.27)
In each case, first identify the redox systems in Table 21.2. Thus, for (i) these are Zn2+I Zn and Mg2+I Mg. Then identify the more negative redox potential. For (i) this is E8(Mg2+I Mg). Then the reducing agent in this system (Mg) is thermodynamically capable of reducing the oxidized state in the other system (Zn2+).Hence Reaction Q.25 is favourable. The same procedure tells you that reactions (ii), (iv) and (v) in the question are also favourable. In the case of (iii), I??(Ca2+ I Ca) is less than E?(Zn2+ 1 Zn), so the favourable reaction is reduction of Zn2+by calcium. This is the opposite of the change in the question, which therefore does not occur.
(i) Aluminium precedes sulfur in Period 3 of Figure 22.1. Consequently, sulfur has the higher ionization energy, because this quantity shows an overall increase across a Period.
(ii) Nitrogen lies above antimony in Group V of Figure 22.1. So nitrogen has the higher ionization energy because this quantity tends to decrease down a Group. (iii) Germanium comes earlier than chlorine in its Period and lower than chlorine in its Group. The two effects of (i) and (ii) reinforce each other: chlorine has the higher ionization energy.
The low first and high second ionization energies of lithium and caesium mean that each atom contributes only one electron to the electron gas, so the metallic bonding is relatively weak. Moreover, as the values of AH&(M, s) in Table 23.1 show, the bonding weakens from lithium to caesium as the radius of the ion core, and therefore the internuclear distance, increases. The rate of reaction depends on the speed with which the metal structure can be broken down. It is therefore greatest at caesium, where the metallic bonding is weakest.
228
Commercial sodium peroxide contains a small amount of the superoxide, Na02, which is formed during the combustion of sodium. The alkali metal superoxides are orange so, at the 10% level, Na02 confers a yellow tint on commercial Na202. In water, this impurity instantly evolves oxygen (cf. Equation 25.4), whereas the preponderant Na202yields H202(aq)according to Equation 25.3. On boiling, the H202formed by both Na202and the Na02 impurity decomposes via Equation 25.5, generating a much larger volume of oxygen.
(a) On heating, the alkali metal hydrides decompose to the metal and hydrogen gas: MH(s) = M(s) + fH&)
As AH: for this reaction is -AH: (MH, s), the values in Table 25.2 show that AH: decreases down Group I from lithium to sodium to caesium. Since AS: should be similar for the three analogous reactions, the decrease in AH: should, according to Equation 17.3, lead to a decreasing decomposition temperature. This is consistent with the facts. As noted in the text, this change occurs because the small size of the hydride ion leads to large decreases in -L(MH, s) as r(M+) increases. (b) With the iodides, the stabilities indicated by the values of reaction MI(s) = M(s) + $ I ~ ( s )
AH: for the
increase down the Group. The iodide ion is large, so when r(M+) increases, the decreases in -L(MI, s) are smaller. Compared with the hydride case, this favours an increase in stability.
The caesium compound is the one to choose because it will be the most stable to the decomposition MClF4(s) = MF(s)
+ ClF3(g)
(Q.29)
which is the reverse of the preparative reaction. This is because the ion ClF4- is obviously bigger than the ion F-. Consequently, -L(MF, s) is larger than -L(MClF4, s), and, in accordance with point 5 of Section 25.4, stability will increase with increasing cation size. In fact, LiC1F4 and NaC1F4 cannot be prepared, but CsC1F4has been made and begins decomposing only at 300 "C.
(a) The immediate neighbours of the magnesium ion are the six oxygen atoms of six surrounding water molecules. This indicates the presence of a magnesium complex, [Mg(H20),l2+. Figure 26.6 tries to convey the fact that the oxygen atoms of the six water ligands are octahedrally disposed around magnesium. (b) MgS04.7H20(s) = [Mg(H20)6]2+(aq)+ S042-(aq) + H20(1)
(Q.30)
(c) According to Table 26.1, the edta4- ligand is hexadentate, so it can replace all six water molecules when a new complex is formed: [Mg(H20)6]2+(aq)+ edta4-(aq) = [Mg(edta)12-(aq) + 6H20(l)
(Q.31)
229
The shortening of the chains suggests that the cavity within the cryptand will be reduced in size. Consequently, the match between ligand and cation size which yields the complex of greatest stability is now more likely to occur at a smaller cation radius. As only Li+ and Na+ have smaller radii than K+, the different cation that forms the most stable complex with cryptand-221 must be Na+; with cryptand21 1 it must be Li+. With cryptand-2 11, the stabilities decrease from lithium to caesium.
Compare the discussion of sodium, magnesium and aluminium in Section 23.1. In terms of the simple electron-gas model of metallic bonding, the Group I1 metals contain M2+ion cores and two electrons per metal atom in the electron gas. This means that the ion cores are pulled together more tightly than the M+ ion cores of the alkali metals. Thus, densities, melting temperatures and enthalpies of atomization are higher for the Group I1 metals.
If, as we surmised with the alkali metals, the rate of reaction depends on how easily the metal structure is broken down (Section 23.1), then the slower reaction of strontium and barium is attributable to their higher enthalpies of atomization.
There are no serious changes in the conclusions. If Ca+ and Rb+ are the same size, then we assume that L(CaC1, s) = L(RbC1, s) = -692 kJ mol-I. Putting this value into the third column of Table 28.4, and adding up the column of figures, AHy(CaC1, s) = -151 kJ mold'. The changes give a value of MF(CaC1, s) that is 26 kJ mol-' more positive than before. Using the estimate AS: = 93.2 J K-' mol-' for Equation 28.8 from Section 28.2, the new estimate for AH? (CaCl, s) gives AGF(CaC1, s) = -123 kJ mol-I. Then, for the reaction (28.1 1) 2CaCl(s) = Ca(s) + CaC12(s) AG:
= -748 kJ mol-' - 2 x (-123 kJ mol-') = -502 kJ mol-I
As before, the calculation suggests that CaCl(s) is stable with respect to its constituent elements, but highly unstable to disproportionation.
The Born-Haber cycle for a difluoride, MFZ, is shown in Figure Q.3 (cf. Figure Q.l). It yields AHF(MF2, S) = AHZm(M, s) + I1(M) + I2(M) + D(F-F) - 2E(F) + L(MF2, s) This means that the top six figures for a fluoride in Table 28.5 must add up to give the bottom number. Thus, in the case of barium, (180 + 503 + 965 + 158 - 656) kJ mol-I + L(BaF2, s) = -1 207 kJ mol-1 whence L(BaF2, s) = -2 357 kJ mol-I
230
Figure Q.3 Born-Haber cycle for the difluoride MF2.
Assuming that Ba2+and Cs2+have the same size, then we take L(CsF2,s) = L(BaF2, s) = -2 357 kJ mol-I. Then, adding up the figures in column 3 of Table 28.5, AHy(CsF2, s) = (76 + 376 + 2 236 + 158 - 656 - 2 357) kJ mol-’ = -167 kJ mol-I Ignoring the AS? term, which will be fairly small, we see that this value suggests that CsF2(s) may be stable with respect to its constituent elements. Its stability is, however, much more likely to be determined by the decomposition to CsF(s), a fluoride in which the cation has a noble gas configuration: CsF2(s) = CsF(s)
+
kF2(g)
(Q.32)
Using our estimated value for AHF(CsF2, s), and the value of M?(CsF, s) in the Data Book: AH: = M?(CsF, S) - AHY(CsF2, S) = -554 kJ mol-’
-
(-1 67 kJ mol-l)
= -387 kJ mol-I
As Equation Q.32 occurs with an increase in the number of moles of gas, AS: will be positive and AGE even more negative than AH:. Thus, our calculations suggest that CsF2 will be very unstable with respect to decomposition into CsF and fluorine. It is therefore not surprising that it has not been prepared.
According to Equation 28.24, each P4O10 molecule is converted into four Po4”ions. Each of the P4O10 molecules shown in Structure 28.1 contains six P-0-P units. In the four Po4’- ions (Structure 28.2), these are replaced by twelve P-Obonds. Thus, two P-0 bonds (in each P-0-P unit) are replaced by two P-Obonds. This parallels what happens in the reactions of CaO with C 0 2 and S O 2 .
The arguments are those developed in Section 25.2. I , and summarized in point 5 of Section 25.4. The Group I1 peroxides decompose on heating as follows: M O ~ ( S=) MO(s)
+
;O,(g)
(Q.33)
The oxide ion, 02-,in the decomposition product, MO, is smaller than the peroxide Thus, decomposition occurs to give a product with a larger value of -L, ion, 022-. and stability increases with cation size. BaO2 will therefore be the most stable of the peroxides, and is most likely to be formed on heating the Group I1 metals in oxygen.
231
The decomposition reaction is MCO~(S)= MO(s) + C02(g)
(28.26)
In Figure Q.4, a thermodynamic cycle has been built around this reaction. This gives: AH: = -L(MC03, S) + L(M0, S) + C -
8W 8W r(M2+)+ r ( C 0 3 2 - ) r(M2+)+ r ( 0 2 - )
+c
Figure Q.4 A thermodynamic cycle for discussing the stabilities of Group I1 carbonates.
As Y ( C O ~ ~is- obviously ) greater than r(02-),a particular increase in r(M2+)will induce a larger decrease in 8W/[r(M2+)+ r(02-)] than in 8W/[r(M2+)+ T(CO~~-)]. Thus, AH: will become more positive as the cation size increases, and stability will increase down the Group. Figure 25.7 covers the essentials of the arguments by using a different example.
In Section 28.1, we related the rates of reaction of the Group I1 metals in water to the solubility of their protective oxide/hydroxide films in the solvent, In water, both beryllium and magnesium are quite well protected, but because B e 0 is amphoteric and MgO is not, the protective film on beryllium is soluble in alkali, and the exposed metal then reacts quickly with water giving hydrogen gas. With the other Group I1 metals, the product would be a solid or dissolved hydroxide, M(OH)2, but the amphoteric nature of Be(OH)2 leads to formation of [Be(OH)4]2-(aq): Be(OH)2(s) + 20H-(aq) = [Be(OH)4]2-(aq) (Q.34) Thus the overall reaction is, Be(s) + 20H-(aq) + 2H20(1) = [Be(OH)4]2-(aq) + H&)
(Q.35)
Two kinds of explanation are possible. Firstly, as electronegativity increases across a Period, the electronegativity of magnesium will be greater than that of sodium, and the electronegativity difference between magnesium and chlorine will be less than that between sodium and chlorine. Secondly, Mg2+is both smaller and more highly charged than Na+, so it is more capable of polarizing the chloride ion. In either explanation, the implication is that MgC12 is less ionic than NaCl, and this is reflected in the structure. 232
The scale appears because equilibrium in dissolution reactions such as CaCO,(s)
Ca2+(aq) + C032-(aq)
(Q.36)
lies well to the left. When [edtaI4-(aq) is added, it removes Ca2+(aq)by forming a complex with these ions. The equilibrium tries to minimize the constraint by supplying more Ca2+(aq).The scale must dissolve to achieve this.
According to Table 26. I , ethylenediamine acts as a bidentate ligand through the non-bonded pairs on its two nitrogen atoms. As beryllium seems to prefer tetrahedral coordination in its complexes, each beryllium will be tetrahedrally bound to two bidentate ethylenediamine ligands in a complex [Be(en)2]2+(Q.l). The chloride will therefore be [Be(en)2]C12.
r
H2 N
2+
233
1 M. Mortimer and P. G. Taylor (eds), Chemical Kinetics and Mechanism, The Open University and the Royal Society of Chemistry (2002). 2 L. E. Smart and J. M. F. G. Gagan (eds), The Third Dimension, The Open University and the Royal Society of Chemistry (2002). 3 C. J. Harding, R. Janes and D. A. Johnson (eds), Elements of the p Block, The Open University and the Royal Society of Chemistry (2002).
Grateful acknowledgement is made to the following sources for permission to reproduce material in this book:
Figures Figure 1.2, 2.1: 0The British Museum; Figure 2.2: Britain on View; Figures 2.4 and 16.2a: Martyn F. Chillmaid/Science Photo Library; Figure 2.8b: Delta Gold; Figure 3.2b: R. Maden, National Geographical Society; Figures 3.6a and 1 2 . 3 ~: Royal Society of Chemistry; Figure 3.6b: Fabre Minerals specimen, F. Fabre; Figure 3.8: Ben Johnson/Science Photo Library; Figure 3.12: photo by Peter Garside, Alcoa World Alumina, Australia; Figure 3.14a: photo supplied by Central Japan Railway Company with cooperation of the Japan Aluminium Association; Figure 3.14b: Mark WagnedFlight Collection; Figure 6.4: Supplied by www.polytechphotos.com; Figure 7.2 and 23.4: Science Photo Library; Figure 8.4: Spencer Swagner/Tom Stack and Associates; Figure 1 2 . I : from the Scientific Papers of J. W. Gibbs. Vol. 1 , Longman Green (1906);Figure 12.1b: Photo RMN-Herve Lewandowski-Louvre; Figure 12.3b: US Department of Energy/Science Photo Library; Figure 2 5 . 1 ~A. : P. Photo; Figure 15.lb: Werner Burgess Fortean Picture Library; Figures 27.I and 28.12: The British Museum; Figure 17.6: Surviving the Iron Age by Peter Firstbrook; Figure 20.5: Godfrey Argent Studio; Figure 20.6: F. Fabre; Figure 21.2: Tek Image/Science Photo Library; Figure 23.2: courtesy of the Wieliszka Salt Mines, Poland; Figure 25.5: V. Yatsina/Novosti; Figure 28.2: Hershel Freidman; Figure 28.3: Britain on View; Figure 28.5: Canadian Space Agency/ Agence Spatiale Canadienne 1996. Received by the Canada Centre for Remote Sensing, processed and distributed by RADARSAT International; Figure 28.15: Jerry SchadBcience Photo Library. Every effort has been made to trace all the copyright owners, but if any has been inadvertently overlooked, the publishers will be pleased to make the necessary arrangements at the first opportunity.
234
Case Study
B a t t e r i e s and fuel cells
Before discussing batteries, we shall revisit electrochemical cells in general, and introduce some new terminology. Electrochemical cells are of two basic types: voltaic or self-driving cells, and electrolytic or driven cells. Self-driving cells were discussed in Section 21 of the main text. In a self-driving cell, chemicals react and generate an electrical current that can be used as a source of electrical power. However, in driven cells, an external source of current is passed through the cell to effect electrolysis and to generate chemicals. In industry, for example, driven cells are used to produce aluminium (Section 3.3), and chlorine and sodium hydroxide as twin products in sodium chloride electrolyis (Section 24.1). Figure 1.1 is a simplified picture of a zinc-chlorine self-driving cell. One electrode consists of metallic zinc deposited on graphite; the other is graphite in contact with chlorine gas. Both electrodes make contact with an acid solution of zinc chloride, ZnC12,containing the ions Zn2+(aq)and C1-(aq). The reaction between zinc and chlorine is thermodynamically favourable. In Figure I . 1, it occurs as two separate electrode reactions. Zinc at the left-hand electrode is oxidized:
left-hand electrode: Zn(s) = Zn2+(aq)+ 2eChlorine gas at the right-hand electrode is reduced: right-hand electrode: C12(g) + 2e- = 2Cl-(aq)
(1.1) (1 -2)
If the two electrode reactions are added, the electrons cancel, and the overall cell reaction is obtained; it is the formation of aqueous zinc chloride solution from zinc and chlorine: Zn(s) + C12(g) = Zn2+(aq)+ 2Cl-(aq) (1.3)
Figure 1.1 A self-driving cell operating in discharge mode: a spontaneous cell reaction produces electric current in an external circuit that bridges the two electrodes. Here, the cell reaction involves the combination of zinc and chlorine. 237
In Figure 1.1, the left-hand electrode has a negative charge because each dissolving zinc atom leaves two negatively charged electrons on the electrode. Likewise, each dissolving chlorine molecule removes two electrons from the right-hand electrode, so the latter is positively charged. The accumulation of like negative charges at the left-hand electrode is dispersed by the movement of electrons from left to right in the external circuit. The flow of current continues until much of the zinc and chlorine has been consumed: the cell has been in discharge mode and is now discharged. But all is not lost. We can now try to start the whole process again by regenerating zinc and chlorine through electrolysis. Figure 1.2 shows how the self-driving cell of Figure 1.1 can be turned into a driven cell. A source of electrical power must be incorporated into the external circuit to drive electrons in the right-to-left direction. In Figure 1.2, this is a powerful battery; the long vertical stroke represents the positive electrode, and the short stroke represents the negative electrode.
Figure 1.2 In a driven cell, a reaction that is not spontaneous is made to occur by the expenditure of energy in a power source inserted into the external circuit. Now the current flow and the reaction are reversed relative to Figure 1.1: zinc and chlorine are produced from aqueous zinc chloride.
Notice that the battery in the external circuit drives the (negative) electrons towards the left-hand electrode. This therefore accumulates negative charge, and is marked negative in Figure 1.2. Likewise, the associated withdrawal of electrons from the right-hand electrode leaves this with a positive charge. Thus, in both Figure 1.1 and Figure 1.2, each type of electrode has the same charge: zinc is negative and the chlorine electrode is positive. The cell is now undergoing electrolysis: the Zn2+(aq)ions are reduced to zinc metal at the left-hand electrode: left-hand electrode: Zn2+(aq)+ 2e- = Zn(s) ( 1-4) The Cl-(aq) ions are oxidized to molecules of chlorine gas at the right-hand electrode: right-hand electrode: 2Cl-(aq) = C12(g) + 2e-
238
(1.5)
The overall cell reaction, the sum of Reactions 1.4 and 1.5 is now: Zn2+(aq)+ 2Cl-(aq) = zn(s)
+ C12(g)
(1 -6)
By itself, this reaction (the reverse of Equation 1.3) is thermodynamically unfavourable. But it has been made to occur by coupling it to the favourable reaction that takes place in the battery in the external circuit: overall, the combination of the two reactions is thermodynamically favourable. The result is that our cell has been recharged with zinc and chlorine: it has been through a charging mode, and is now ready to be used again as a self-driving cell to generate electrical power. In reading about batteries, you will often see electrodes described as anodes or cathodes. Whether the cell is in charge or discharge mode, the anode is the electrode where oxidation takes place, and the cathode is the site of reduction. In Figures 1.1 and 1.2, identify the anode and cathode. In Figure 1.1, oxidation occurs on the left, zinc electrode, and reduction at the right, chlorine electrode. Thus, in Figure 1.1, zinc is the anode and chlorine is at the cathode. In Figure 1.2, reduction occurs at the left and oxidation at the right. Zinc is now the cathode, and chlorine is at the anode. We have already noted that in both the self-driving (discharge) and driven (charge) modes, the sign of a particular electrode is the same. Thus, in both Figures 1.1 and 1.2, zinc is negative. So, in switching modes, the sign of a particular electrode is unchanged, but the cathode and anode change places.
239
A battery consists of a self-driving cell or cells. The zinc-chlorine system that we looked at in Section 1 is potentially an example of a secondary battery: it can be recharged many times, and for this reason has been explored as a possible power plant for electric vehicles. But most batteries that you use are not like this. They are primary batteries, which are discharged once and then thrown away. The most important attribute of a battery is portability, allowing electrical devices to be operated away from a mains supply of electricity. A battery has certain essential components. The negative electrode consists of a metallic current collector or ‘terminal’ and an active component, which takes part in the electrode reaction. The latter is often a finely divided metal such as zinc, lead, cadmium or lithium, which can be oxidized with the release of electrons. The positive electrode also consists of a metallic current collector and an active component, generally a metal oxide (e.g. Mn02, Pb02, NiO(OH), Ago), in which the metallic element is in a highly oxidized state. For example, if NiO(0H) is thought of as a combination of 02-,OH- and Ni3+,then nickel is in oxidation state +3. In a cell, NiO(0H) is often reduced to Ni(OH)2, in which nickel has its more common, lower oxidation state of +2. The electrodes are separated by an electrolyte, which conducts ions, but which must be an electronic insulator to avoid internal short circuits. In most conventional batteries the electrolyte is an aqueous solution such as ZnC12, KOH or H2S04,although some advanced batteries use organic electrolytes, ion-conducting ceramics, polymers or molten salts. The positive and negative electrodes of a battery cell are generally placed close together to minimize the internal resistance of the cell (which leads to loss of voltage when drawing a current). To prevent the electrodes touching inadvertently, so causing an internal short circuit, a mechanical separator is employed, and this is another essential cell component. Often it is in the form of a microporous sheet of polymer to hold the electrodes apart, but which absorbs electrolyte and permits ions to pass. These separators make little contribution to the cell resistance and mass. For some applications, such as electric cars, a multi-cell battery consisting of a network of identical cells may be required. In this network, some cells may be connected in series as in Figure 2.la; this increases the voltage. Others may be connected in parallel (Figure 2. lb.); here, the voltage of the combination is only that of a single cell, but the time for which a particular level of current can be delivered is much increased. It is important that the voltages of parallel cells remain identical after installation. Since the cell voltage changes somewhat during a cell’s lifetime, it is advisable that parallel cells should be of comparable age. To see why, consider any two of the three cells in Figure 2.1b. They are connected in the same sense as are the battery and driven cell of Figure 1.2. If one has a higher voltage than the other, it will discharge, and charge the cell of lower voltage, as in Figure 1.2. So, if parallel cells are not well matched in voltage (as a result of a variation in their ages), the cells of lower voltage will drain current from their neighbours. Thus, in a large multi-cell battery, like the emergency power supply in a building, it is not uncommon for internal currents to circulate through parallel cells even when no external load is connected. 240
Figure 2.1 cellsconnected (b) in parallel.
,ca)
in series; and
Over the past 50 years the applications for small sealed batteries in the home have expanded phenomenally. Today, small primary or rechargeable batteries are employed in a huge number of appliances. Some examples (Figure 3.1) are as follows. Household: telephones, clock-radios, security alarms, smoke detectors, portable fluorescent lamps, torches and lanterns, door-chimes, car central-locking activators. Workshop and garden: portable tools (e.g. screwdrivers, drills, sanders), portable test meters, hedge trimmers, lawnmowers.
Entertainment: portable radios and televisions, compact disc players, tape recorders, remote controllers for televisions and videos, electronic games and toys, keyboards. Personal hygiene and health: toothbrushes, bathroom scales, hair trimmers, shavers, blood-pressure monitors, hearing aids, heart pacemakers. Portable electronic devices: watches and clocks, cameras, camcorders, calculators, organizers, mobile phones, laptop and notebook computers, bar-code readers.
Figure 3.1 A range of batteries of different shapes and sizes. 24 I
Many of these applications require batteries of advanced design that give greatly improved performance as a result of developments in materials science and technology. Although most small consumer batteries are still of the primary (‘throw-away’) variety, there is a growing trend to adopt rechargeable batteries because they are more economical. Moving to larger secondary batteries, principally lead-acid, there has been a growth in the market for engine-starter batteries to mirror the growth in transport. Every internal combustion engine, whether it be in a car, a truck, a bus, an aircraft or ship, requires a starter battery. These were once ‘starting, lighting, ignition (SLI) batteries’, reflecting their principal functions. With vehicles becoming more sophisticated, the need for electric motors and other electrical facilities has mushroomed. The demands on the battery have grown correspondingly, and it may soon be necessary to fit two batteries to the private car; one for engine starting, the other for running auxiliaries. At the same time, ‘SLI batteries’ have become ‘automotive batteries’ to reflect the other functions of the modern car battery. There has been a similar rise in demand for very large, installed battery packs. Almost every public building (airports, hospitals, hotels, railway stations, stores and supermarkets, etc.) must have an uninterruptible power supply (UPS), so a battery pack can take over seamlessly when the mains supply fails, until such time as a local generator can be started. Other applications are shown in Table 3.1. In the past decade, new rechargeable batteries have been introduced into civilian and consumer markets. Moreover, there are now specialized batteries with unconventional chemistries for use in the military and the space fields. As Table 3.1 shows, the size range of batteries is now enormous. The energy content of a battery is expressed in watt hours (W h). As 1 W = 1 J S-I, and 1 h = 3 600 s, 1 W h = 3 600 J. The scale extends from small button cells of energy content -0.1 W h, all the way to load-levelling batteries of energy content -10 MW h. Few other commodities come in such a range of sizes. Some typical applications for batteries of differing size are also given in Table 3. I .
Table 3.1 Battery sizes and applications
242
As the current that a battery delivers increases, its voltage tends to fall. To understand this, we return to the cell considered in Section 21 of the main text. The method described there for determining a voltage for this cell is shown in Figure 4.1. An adjustable counter-voltage is inserted into the external circuit. When the counter-voltage, V, j u s t stops the current generated by the cell, its value is equal to the cell voltage, E. The cell voltage, E, obtained at the point of balance is often called the electromotiveforce, or e.m.f. of the cell.
Figure 4.1 Measuring the voltage or e.m.f. of a reversible electrochemical cell. When the counter-voltage V in the external circuit is less than E, the e.m.f. of the cell, hydrogen gas reduces copper ions, and electrons flow from left to right in the external circuit. When V is greater than E , electrons flow from right to left, copper is oxidized, and hydrogen ions are reduced. Such a cell is said to be reversible. When E and V are equal, there is no current and no reaction.
In Section 21 of the main text, we also said that the reversible, zero-current voltage, E, is the maximum possible voltage that the cell can deliver. When the cell is discharging, and delivering current to the wire, its discharge voltage, is less than E . This is serious. It is useful to think of the voltage as the energy supplied to the external circuit per passed electron. If the voltage drops, the energy per electron drops. Discharge of the cell therefore means loss of energy that might have been used to do useful work by, say, driving an electric motor placed in the external circuit. Why does current flow cause this? Notice first that whatever the size of the current, the energy liberated by a given amount of reaction in the cell is the same. The difference is that when a substantial current is flowing, less of this energy is available to do work. If it is not expressed as work, it must appear as heat. So the drop in the voltage when a cell delivers a significant current is associated with a greater evolution of heat. 243
One such source of heat is the internal resistance of the cell. The current flowing in the external circuit must also pass through the cell. There, it encounters resistance, just like a current flowing through the resistance of the wire heating element in an electric fire. The result in both cases is evolution of heat. In the case of the cell, this is initiated within the electrolyte. There are also heat losses at the cell electrodes. Suppose Figure 4.1 is brought to the zero current situation with a stopping voltage. Now we reduce the stopping voltage very slightly so that a tiny current flows in the wire. The current is so small that the polarity of the electrodes is not disturbed. Here, the discharge voltage, Vd, is effectively equal to E. Now we lower the counter-voltage by a much greater amount, so that a substantial current flows in the wire. The flow of electrons out of the negative (left-hand) electrode tends to reduce the density of the negative charge; the flow of electrons into the right-hand electrode tends to reduce the density of positive charge. These sudden losses cannot be entirely made good by adjustments that they prompt elsewhere in the cell. So the initiation of a significant steady current diminishes the polarity difference between the two electrodes. This effect, like that arising from the internal resistance of the cell, tends to make the discharge voltage, Vd, less than E. The greater this deficiency, the greater the evolved heat. Such losses of voltage arise from changes of polarity. They are called polarization losses, and they can occur at both electrodes, and also in the electrolyte. Polarization losses arise because when a steady current is started, the response of the resulting chemical changes initiated within the cell is never fast enough to restore the zero current polarities. For example, in a typical electrode, such as the nickel oxide positive electrode employed in rechargeable alkaline-electrolyte batteries, the reversible discharge reaction may be shown as: NiO(OH)(s) + H20(1) + e- 6 Ni(OH)2(s) + OH-(aq) (4.1) In the left-to-right reaction, Ni3+is being reduced to Ni2+.For this process to take place, an electron from the current-collector has to interact with a water molecule and a particle of the solid nickel oxy-hydroxide. On a macroscopic scale, this will begin with particles of solid immediately adjacent to the current-collector, since these are encountered first by the emerging electron. As discharge proceeds, a reaction front will propagate into the bulk of the active material and away from the current-collector. Most batteries perform better if allowed to ‘rest’ or ‘recuperate’ periodically during discharge. This permits diffusion to take place, which reduces concentration gradients in the solids taking part in the battery reaction, and restores a balanced situation, nearer to equilibrium. On charging the battery, the reverse processes take place, again with diffusion controlling the reaction paths. The cell voltage, therefore, drops increasingly below the reversible, zero-current value as the current drawn from it is increased. Figure 4.2 shows how this and similar effects work in practice for the charge and discharge of a small sodiummetal-chloride cell. The horizontal axis, with the unit ampere hours (Ah), shows the charge that remains available in the cell at different stages of the discharge/ charge cycle. In SI units, charge is measured in coulombs (C), and 1 A h = 3 600 C. The discharge voltages (lower blue curves) lie below the reversible value of 2.35 V. At 9 Ah, after 1 A h has been drawn from the cell, the discharge voltages decrease from about 2.25 V when the current is 5 A, to about 1.51 V when the current is 20 A. Each discharge curve also shows that, if a constant current is drawn from the cell, the voltage falls as the charge still available in the cell decreases; that is, the cell voltage decreases with time. 244
Figure 4.2 The charge remaining available in a sodium-metal-chloride cell at different stages in the dischargekharge cycle. It is plotted against the cell voltage for a range of constant currents. The lower blue discharge curves should be read from left to right; the upper yellow charging curves from right to left.
Notice that whereas the discharge voltages lie below the reversible value, the voltages for the charging curves lie above it. On discharge, the effects of polarization and internal resistance cause a voltage drop. During charging, such effects have to be overcome, and the required voltage exceeds the reversible value. Ampere hours is also the unit used as a measure of the storage capacity of a cell or battery. This is the product of the number of hours for which the cell may be discharged at a constant current to a defined cut-off voltage at which the cell is no longer very useful. The value of the capacity depends not only on the ambient temperature and on the age/history of the cell, but also, to a greater or lesser extent, on the rate of discharge employed. The higher the rate of discharge, the less the available capacity. At the same time, as we have seen, the operating voltage falls off markedly at high discharge rates because of polarization. The result is that the stored energy delivered, measured in watt hours (W h = V x Ah), declines even more sharply at higher discharge rates. When considering the capacity and the stored energy available from a battery, it is essential to define both the discharge rate and the temperature to be employed. Battery manufacturers generally state a rated capacity (so-called nominal or name-plate capacity) under specified discharge conditions, often the 5-h rate at 25 "C to a designated cut-off voltage. This can be determined by finding the constant current that causes a fall to the cut-off voltage after exactly 5 hours.
245
The specification for a battery depends very much on its proposed use. There are three broad considerations: technical performance, environmental factors (safety and recycling or disposal), and cost. Technical performance is affected by cell chemistry, materials of construction, design, ambient temperature, rate of discharge, depth of discharge before recharging, cut-off voltage on recharge, etc. We list below some of the desirable attributes looked for in a battery. All batteries
high cell voltage and stable voltage plateau over most of the discharge high stored energy content per unit mass (W h kg-') and per unit volume (W h litre-l) low cell resistance (milliohms) high peak power output per unit mass (W kg-') and per unit volume (W litre-') high sustained power output wide temperature range of operation long inactive shelf life (years) long operational life low initial cost reliable in use sealed and leak proof rugged and resistant to abuse safe in use and under accident conditions made of readily available materials which are environmentally benign suitable for recycling
Secondary batteries For these we may add: high electrical efficiency (W h output/W h input) capable of many charge/discharge cycles ability to accept fast recharge will withstand overcharge and overdischarge sealed and maintenance-free This is a formidable list, and compromises have to be accepted. The relative importance of the various factors varies widely according to the application. For example, reliability under all conditions of use is relatively unimportant for an automotive battery; if it fails prematurely it is merely an inconvenience and a cost consideration. Contrast this with an aircraft battery, where sudden failure in flight would be much more serious, or with a battery powering a satellite, where -there being no access to the faulty item -failure would result in the loss of a multimillion pound vehicle. 246
A further complication is that the above desirable factors are often highly interactive. For example, the available stored energy and the peak power output both depend on the temperature; the peak power also depends on the state of charge of the battery; the charge-discharge cycle life of a secondary battery depends critically on the severity of the discharge in each cycle, and so on. All these factors need to be quantified before we can decide whether a battery is likely to be commercially viable for a particular application.
By far the most common primary cells are based on the zindmanganese dioxide system, either so-called ‘zinc-carbon cells’ (Leclanchk cells) or alkaline manganese dioxide cells. These both give 1.5 V open circuit, but differ in a number of important respects. Zinc-carbon cells (Figure 5.la) have a central carbon rod immersed in the positive active material, or cathode (a mixture of compressed impure Mn 0 2 and carbon), a container of metallic zinc as the negative electrode, or anode, and an electrolyte of aqueous NHdC1 and/or ZnC12 These cells are traditional and inexpensive. The chemistry is complex because there are several competing cell reactions. Here we give just one of them. At the anode, zinc undergoes its usual oxidation to Zn2+;at the cathode, the oxidation state of manganese is reduced from +4 in Mn02 to +3 in MnO(0H): Mn02(s) + H20(1) + e- = MnO(OH)(s) + OH-(aq) (5.1) As zinc hydroxide, Zn(OH)2, is almost insoluble in water, the zinc and hydroxide ions combine to precipitate this solid, and the overall reaction is: Zn(s) + 2Mn02(s) + 2H20(1) = Zn(OH)2(s) + 2MnO(OH)(s) (5.2)
Figure 5.1 (a) Zinc-carbon battery; (b) alkaline manganese dioxide battery. 247
Alkaline manganese dioxide cells (Figure 5 . lb), a superior and more expensive product, use finely divided zinc powder as the negative electrode, and this fills the centre of the cell, with a brass pin to make contact with the base. The electrolyte is concentrated KOH solution, and the positive electrode material, a mix of chemically or electrochemically prepared Mn02 and carbon, forms a concentric annulus around the zinc powder and the separator. Alkaline manganese dioxide cells have a long shelf life and are particularly useful for high-drain (power) applications, where their useful life is several times that of zinc-carbon. The cheaper zinc-carbon cells are adequate for low-drain applications and for intermittent use (such as in torches), where there is recovery time between uses to allow diffusion processes to remove polarization at the electrodes and restore equilibrium. Both types of cell are made by most manufacturers in a variety of standard sizes and shapes. The prismatic (rectangular) 9 V batteries, as used in smoke detectors, contain six small cells connected in series. Higher voltages can be obtained by using cells containing anodes made of metals whose oxidation is thermodynamically more favourable than that of zinc. Lithium (see Tables 12.1, 15.1 and 21.2 of the main text) is one such candidate, and several manufacturers are now offering 3 V lithium-Mn02 cells. Water cannot be used in the electrolyte because lithium reacts with it, so these cells employ a lithium foil negative electrode and an ion-conducting organic electrolyte. The latter consists of a lithium salt, such as LiAsF6 or LiCF3S03,dissolved in a polar organic solvent such as 1,2-dimethoxyethane (Structure 5.1), or tetrahydrofuran (Structure 5.2). They are available as cylindrical cells, using spirally wound electrodes (so-called ‘jelly roll’ configuration, Figure 5.2a), or as miniature (button or coin) cells. In spirally wound cells, the lithium anode in the form of sheet foil is laid flat; a separator sheet is superimposed on it, and finally there is a sheet of bonded Mn02 serving as the cathode. The three layers are laminated by passing through rollers under pressure
H3COCH2CH20CH3 dimethoxyethane 5.1
tetrahydrofuran
Figure 5.2 (a) High-rate lithiudmanganese dioxide primary cell, spirally wound; (b) the commercial product. 248
5.2
and then cut to size and rolled into a cylindrical 'Swiss roll,, which just fits in the can. Because the component layers are thin, and the rolling is tight, there is a far greater surface area of electrode in this type of cell than in a conventional alkalinemanganese cell, resulting in lower internal resistance. This is needed because the ionic conductivity of the organic electrolyte is much lower than that of concentrated KOH solution. As lithium is such a powerful reducing agent, with a density only half that of water, lithium batteries deliver relatively high energies per unit mass. They also have a long shelf life and the ability to operate over a wide temperature range (-40 OC to + 60 OC). Lithium button and coin cells are widely used in watches, clocks and pocket calculators. Among the possible cathodes are Mn02 or a polymer with an empirical formula close to CF. This can be produced by the action of fluorine on graphite. In CF (Figure 5.3), the graphite rings have become puckered, each carbon forms one C-F bond, and the multiple bonding in the rings has been eliminated. Fluorination to a composition CF,, where x is 0.5-0.95 leaves enough multiple bonding in the rings to provide some conducting properties, and the polymer can then be used as a cathode. If we write it as (CF0.8)n,the cathode reaction is: CnFO,gn(s)+ e- = CnFO.gn- I(S)
+ F-(soIv)
(5.3)
But lithium fluoride is insoluble in the organic solvent, so when fluorine enters it as fluoride ion, it combines with Li+(solv) to give solid lithium fluoride. The overall cell reaction is therefore: Li(s) + C~zF0.8n(~) = LiF(s) + c n F 0 . 8 ~ -1 l(s)
(5.4)
Figure 5.3 Structure of the layers of graphite after fluorination to the composition (CF),. When fluorinated graphite is used as a battery cathode, the degree of fluorination is reduced. 249
Alternative systems of lower voltage include zinc-manganese alkaline (1.5V) and zinc-silver oxide cells (1.5 V). Altogether, there are over 40 different sizes and chemistries of button and coin cells. In the military field, large batteries have been used to propel torpedoes. The MU-90 torpedo (Figure 5.4), developed for the French and Italian navies, relies on an aluminium-silver oxide cell with an alkaline electrolyte. Al(OH)3 is amphoteric, and in alkali dissolves to form [A1(OH)4]-(aq) (Section 3.3 of the main text). The silver oxide cathode consists of A g o in which silver is in oxidation state +2. The cell reaction is: 2Al(s) + 20H-(aq) + 3H20(1) + 3AgO(s) = 2[Al(OH),] -(aq) + 3Ag(s) (5.5) The guidance systems of missiles are powered by large, lithium primary batteries, which use a molten lithium salt as electrolyte. The battery operates at >300 O C , and is heated rapidly to operating temperature by a pyrotechnic charge.
5.2.I Lead-acid batteries By far the most widely used rechargeable battery is the lead-acid battery (Figure 5.5). This finds applications in engine starting (automotive batteries), supplying domestic electrical loads for caravans and boats (leisure batteries), standby power supplies for hotels, shops, hospitals, factories, etc. (stationary batteries), for powering electric vehicles (traction batteries), and as back-up for solar and wind-powered installations. It differs from most other rechargeable batteries in that the electrolyte (sulfuric acid) participates in the electrode reactions, and is more than just an ionic conductor. This is made clear in the following electrode reactions:
negative electrode Pb(s) + H2S04(aq)= PbS04(s) + 2H+(aq) + 2e-;
= 0.356V
(5.6)
positive electrode Pb02(s) + H2S04(aq) + 2H+(aq) + 2e- = PbS04(s) + 2H20(1); E@ = 1.685 V overall cell reaction: Pb(s) + Pb02(s) + 2H2S04(aq) = 2PbS04(s) + 2H20(1);
= 2.041 V
Figure 5.4 The battery-powered MU-90 European torpedo.
(5.7)
(5.8)
During discharge, sulfuric acid is consumed and water is formed, with the converse on charging. The density of the electrolyte varies according to the density of sulfuric acid, which allows the density of the electrolyte to be used as a measure of the state of charge of the battery. Although the lead-acid battery dates back to the nineteenth century, many improvements in its design and construction have been made, and, indeed, are still being made. Many rely on the use of new materials for the 'inactive' components (current collectors, separators, seals, battery cases, etc.) and on new manufacturing technology. Consequently, modem lead-acid batteries are greatly superior in performance to those of fifty years ago, and, in real terms, they are cheaper. Their principal limitation still lies in their excessive weight, a consequence of the high density of the solid lead. This places a fundamental limitation on the specific energy (W h kg-I) that may be achieved with a lead-acid battery. Most practical lead-acid batteries deliver just 30-40 W h kg-'. 250
Figure 5.5 The familiar lead/acid starting battery found beneath nearly every car bonnet.
5.2.2 Alkaline batteries Another well-known rechargeable battery dating back 100 years is the nickelcadmium battery. These are commonly found in small rechargeable appliances such as toys, portable cassette players and toothbrushes (Figure 5.6), as well as in much larger sizes for use in aircraft and as traction batteries for electric vehicles. The large ones are superior to lead-acid batteries in many respects, in particular their long life of up to 2 000 charge-discharge cycles, but they are also much more expensive.
Figure 5.6 The CD player, and the electric toothbrush are two common devices often powered by rechargeable nickel-cadmium batteries.
The nickel-cadmium battery is an example of a class of rechargeable battery with an alkaline (KOH) electrolyte. The electrode reactions are:
negative electrode Cd(s) + 20H-(aq) = Cd(OH)2(s) + 2e-; @ = 0.81 V positive electrode 2NiOOH(s) + 2H20(1) + 2e- = 2Ni(OH)*(s) + 20H-(aq);
(5.9) = 0.49V
(5.10)
overall reaction Cd(s)
+ 2NiOOH(s) + 2H20(1) = Cd(OH),(s) + 2Ni(OH),(s);
= 1.30V
(5.11)
At 1.30 V, the voltage of this battery is only two-thirds that of the lead-acid battery, and cadmium ih also a relatively dense metal. As a result of these two factors, the specific energy of the nickel-cadmium battery is little better than that of the leadacid cell. However, the nickel-cadmium type has other more marked advantages. Besides the longer cycle life already noted, it has a flatter discharge voltage, better low temperature performance, a continuous overcharge capability without causing damage, excellent reliability and low maintenance requirements. In recent years a new alkaline nickel oxide battery has been developed and is now widely used in, for example, mobile phones (Figure 5.7). This is the so-called ‘nickel-metal hydride’ battery, usually abbreviated as NiMH. This has the same nickel oxide positive electrode and the same KOH electrolyte as a nickel-cadmium battery, but a new negative electrode in the form of a metal hydride. The negative electrode reaction is MH,(s) + OH-(aq) = MH, - I(s) + H20(1) + e(5.12) 25 I
and the positive electrode reaction is the same as for the nickel-cadmium battery. The overall cell reaction is then: (5.13) NiO(OH)(s) + MH,(s) = Ni(OH)2(s) + MH,- l(s) The practical cell voltage lies in the range 1.2-1.3 V, which is almost the same as nickel-cadmium batteries, and the two batteries are interchangeable. The metal hydride in this battery, which has been specially developed and optimized for it, is formed from a complex alloy of several metals, based either on lanthanide elements and nickel, or on titanium-zirconium. The specific energy of N iMH batteries is 60-70 W h kg-’, almost twice that of Ni-Cd. Also, they are resilient to overcharge and over-discharge, have a high specific power output, and may be used over a wide temperature range (-30 to +45 “C). Given the toxicity of cadmium, which poses an environmental risk when disposing of used batteries, it seems likely that metal hydride batteries will progressively take over from cadmium batteries for many applications.
5.2.3 Lithium batteries The desirable qualities of lithium batteries were mentioned in Section 5.1. Among the metals, it is the least dense and one of the most reactive. These qualities make it an obvious candidate for battery use, providing a much higher electrochemical reduction potential than zinc; its batteries have specific energies of some 130 W h kg-’. The problems in developing lithium batteries stem from the high reactivity of lithium metal, particularly with water. This necessitates using a dry room for handling the metal and constructing cells. It is also necessary to use a nonaqueous electrolyte, which may be either an organic liquid or a solid polymer, each with a dissolved lithium salt to make it ionically conducting. As already mentioned, primary lithium batteries using a lithium foil negative electrode, an organic liquid electrolyte, and any one of several positive electrode materials are commercially available. The positive electrodes used include Mn02, MoS2, V6OI3and a ‘carbon fluoride’ CF, (Figure 5.3). The difficulties arise when one tries to develop a rechargeable lithium battery of this type. This involves the electroplating of lithium metal from an organic electrolyte. Much work has been done in this field, with only limited success. Lithium is not normally electrodeposited as a smooth layer on the metal current collector, but as a ‘mossy’ deposit. Lithium foil that has been exposed to air is covered with a thin layer of hydroxide and nitride, which offers some measure of protection from further corrosion. Freshly electrodeposited lithium is finely divided and highly reactive, and decomposes the electrolyte. This is highly detrimental, and can be dangerous if it leads to build-up of gas pressure within the cell. Some of the metal deposit becomes electrically isolated from the electrode, and so capacity is lost rapidly. Lithium metal may also plate out in the form of tree-like crystals known as ‘dendrites’. These can ultimately penetrate the separator, bridge the electrodes, and cause an internal shortcircuit of the cell. Cells have been known to ignite spontaneously during recharge. For all these reasons, the commercial prospects for rechargeable cells based on liquid electrolytes and lithium metal negatives are not too promising, although much research is still in progress, particularly using solid polymer electrolytes. It is the lithium-ion battery, the rising star of recent years, which circumvents most of these problems. The essential feature of the lithium-ion battery is that at no stage in the chargedischarge cycle should any lithium metal be present. Rather, lithium is incorporated 252
Figure 5.7 Many mobile phones use rechargeable
nickel-metal hytlride batteries.
(‘intercalated’) into the positive electrode in the discharged state and into the negative electrode in the charged state, and it moves from one to the other through the electrolyte as Li+ ions. The electrolyte is a solution of a lithium salt in an organic solvent. Figure 5.8 shows a schematic diagram of a secondary battery of this sort.
Figure 5.8 A schematic diagram of a lithium ion ‘rocking-chair’ battery. The item marked ‘lithium’ in the key represents the sites between which lithium ions move during charge and discharge. On discharge, lithium moves out from between the layers of the carbon atoms in the graphitic electrode as Li+ ions. These pass through an electrolyte consisting of a lithium salt dissolved in an organic solvent or polymer. They then enter an oxide of a transition metal such as cobalt at the other electrode, and become bound to oxygen. In charging mode, lithium ions move back from the oxide to the graphite.
The origin of the cell voltage is the difference in the Gibbs function of lithium in the crystal structures of the two electrode materials. Commercial cells use carbon as the negative electrode because the Gibbs function change for the reaction of lithium metal with graphite is relatively small. The lithium in this material therefore retains most of its thermodynamic strength as a reducing agent. Lithium will intercalate readily into graphite up to a composition approaching C6Li. This material is a metallic alloy. If we use a simple electron-gas model of metallic bonding, one can envisage the intercalated lithium as Li+ ions lying between the planes of carbon atoms. As you know, lithium undergoes a very energetic reaction with oxygen to form the oxide, Li20 (Table 15.1 of the main text), in which lithium ions are surrounded by oxide ions. So positive electrodes that can supply an oxide environment for lithium ions are very suitable for lithium batteries. Such electrodes are normally based on 253
either cobalt oxide or nickel oxide. The cells are assembled in the discharged state using a positive electrode of LiCo02 or LiNi02 from which Li+ ions are ‘de-intercalated’ on charging, thereby transferring them into the carbon negative electrode. A 3 V lithium-ion cell results. After a few initial charge-discharge cycles, approximately one-half of the intercalated lithium may be removed reversibly from the positive electrode, as shown: charged discharged (5.14) Lio,ssCo02+ 0.45Li+ + 0.45e- +LiCo02 (1 23 A h kg-l) Li0,35Ni02+ 0.5Li+ + 0.5e-
Li0,8sNi02(135 A h kg-l)
(5.15)
The values in brackets give the practical electrode capacity per kilogram of positive active material for the de-intercalation of the amounts of Li+ shown.
Lithium-ion cells are constructed in the same ‘spiral wound’ configuration as primary lithium cells (Figure 5.2). When cycling Li+ ion cells, it is important to control the top-of-charge voltage carefully (4.1 V for LiNi02 and 4.2 V for LiCo02). Failure to do so results in decomposition of the positive electrodes to give oxygen gas and Co304or LiNiz04,a hazardous situation in a sealed cell. For this reason, lithium-ion cells must be recharged using a specially designed charger incorporating both voltage and temperature control. Over-discharge must also be avoided, and it is usual to have a limiting cut-off voltage on discharge of ca 2.7 V. (In this regard the NilMH battery has the advantage of being much better able to withstand overcharge and overdischarge). Much research is in progress on the possibility of using a lithium-manganese oxide as the positive active material in place of LiCo02 and LiNi02. This would lead to cost reductions, since manganese is much cheaper than cobalt or nickel. Unfortunately, the structure of the lithium-manganese oxide system is less easy to control, and the development of a fully satisfactory lithium-manganese oxide electrode has not proved to be easy.
The SONY Corporation in Japan first commercialized lithium-ion cells in the early 1 9 9 0 and ~ ~ other battery manufacturers now also make them. They are widely used in Iaptop computers (Figure 5.9), mobile phones and other portable electronic equipment. In 1998 alone, about 280 million cells were manufactured in Japan. With much research and development in progress, we can expect future improvements in performance as well as price reductions.
Figure 5.9 The high specific energy of rechargeable lithium batteries accounts for their frequent use in laptop computers.
254
If a battery is to be charged/discharged for hundreds, or even thousands, of cycles, it is essential that the chemical reactions taking place at the electrodes are quantitatively reversible. Even if as little as 0.1% irreversibility (or side-reaction) occurs, this will soon add cumulatively to a major loss in capacity. Many, if not most, electrode reactions involve a reconstructive phase change in the crystal chemistry of the active materials. A typical positive electrode reaction would be solid A + anion e. solid B + e(6-1) (e.g. Ni(OH),(s) + OH-(aq) +NiOOH(s) + H20(1) + e-) This involves ionic diffusion processes in the crystal structure of the solids, leading to phase change and recrystallization. The need to reverse this reaction quantitatively during each cycle is a very demanding one. The severity of the specification is apparent when one considers the many possible processes or sidereactions leading to battery deterioration and failure. These include: growth of metallic needles at the negative electrode, causing internal short circuits ; mechanical shedding of active material from electrode plates; separator dry-out through over-heating; corrosion of current collectors, resulting in increased internal resistance; gas formation at electrode plates on overcharge, causing disruptive effects. These and other degradation processes may result in sudden battery failure, through an internal short circuit, or may lead to progressive loss in capacity and performance. Generally, the degradation steps are interactive and accumulative, so that, when the performance starts to deteriorate, it soon accelerates and the battery becomes unusable. Nevertheless, remarkable success has been achieved in designing batteries of long cycle life (-1 000 cycles) for several different chemistries. Our modern telecommunications, weather forecasts and military defences are totally dependent on the performance of such batteries in satellites, where lifetimes of > 20 000 cycles are sometimes realized. Figure 6.1 shows one of the more recent examples, for which a nickel-metal hydride battery has been chosen.
Figure 6.1 The Optical Inter-orbit Communications Test-Satellite (OICETS) is a joint European and Japanese venture. Its nickel-metal hydride batteries are recharged from the solar panels, here shown fully deployed. 255
Fuel cells are essentially water electroly sers working in reverse; a fuel gas (normally hydrogen) is fed to one electrode and oxygen (or air) to the other. Electrochemical reaction takes place to form water, generating a voltage across the cell, with the fuel electrode being negative and the oxygen electrode positive. A direct current may then be drawn. A schematic diagram of a fuel cell is shown in Figure 7.1.
Figure 7.1 Schematic diagram of a fuel cell.
The fuel cell was invented in 1839 by Sir William Grove, and has been much investigated ever since on account of its perceived advantages for power generation. Compared with the internal combustion engine, fuel cells have high thermodynamic efficiency, and good performance under low load conditions. Other advantages include low pollution emission, rapid response time, simplicity of mechanical engineering and modular factory construction (which facilitates the gradual buildup of large units). It has, however, proved difficult to turn fuel cells into commercially viable power devices, using natural gas, oil or coal as primary fuels. Such fuels must first be ‘reformed’ (i.e. converted) to hydrogen, and the difficulty of developing a satisfactory reformer is comparable to the difficulty of developing the fuel cell itself. There is also the problem of poisoning of the catalyst used in the fuel reformer, and of the electrocatalyst in the fuel cell, by impurities (especially sulfur) in the primary fuel. Altogether, a fuel cell that uses primary fuels poses complex engineering problems. 256
With pure, electrolytic hydrogen there is no need for a reformer and no impurities to worry about. The problem of developing a satisfactory fuel cell is therefore much simpler. Hydrogen-oxygen fuel cells have been used in manned space flight, the product water being drunk by the astronauts. Plans have also been developed for combined electrolysers and fuel cells for this application, as a means of storing solar electricity generated by photovoltaic cells. Thus, space technology is providing a lead for future terrestrial applications. A fuel cell differs from a conventional battery in that the reactants are gaseous and stored outside the cell. Therefore the capacity of the device is limited only by the size of the fuel and oxidant supply. For this reason, fuel cells are rated by their power output (kW) rather than by their stored energy (kW h). There is interest in fuel cells for both static (e.g. combined heat and power generation for buildings) and mobile applications. A particular attraction of fuel cells for mobile applications -for example, powering electric vehicles -is that the positive reactant (air) does not have to be transported, and is free. Unfortunately, this advantage is offset by the difficulty of conveying hydrogen. Compressed hydrogen in gas cylinders is both bulky and heavy, liquid hydrogen is impractical for most uses, and hydride storage beds are still in experimental stages. So a fuel-cell powered car would be likely to use liquid methanol as a fuel, with a reformer on board to decompose it to hydrogen and carbon dioxide. This concept is now in an advanced development stage, with several major automotive companies (Daimler-Benz, Ford, etc.) involved. For larger vehicles, buses and trucks, it is possible that compressed hydrogen gas in cylinders might be used. Ultimately, the oil companies would favour the development of an on-board reformer for petroleum. Fuel cells come in a number of different types, which differ in the electrolyte they employ and the temperature range over which they operate. The thermodynamic reversible voltage for the decomposition of water, calculated from the values of the Gibbs function, decreases linearly from E? = 1.229 V at 298 K to 1.088 V at 473 K. Thermodynamics therefore favours the operation of a water electrolyser at relatively high temperature. In contrast, a fuel cell, being the reverse electrochemical reaction, is favoured by operation at low temperature and high pressure, but kinetic factors (electrocatalysis, polarization) call for higher temperatures and a compromise must be made. At high temperatures, the overall energy efficiency can be maintained by waste heat recovery, even though the net voltage efficiency is lower. Typically, fuel cells generate only 0.7-0.8 V. The five principal types of fuel cell are summarized in Table 7.1. These have been developed for different applications. Phosphoric acid fuel cells operating at 200 "C are now manufactured in large (MW) unit sizes to supply combined heat and power (CHP) to large building complexes. Alkaline fuel cells, used first in spacecraft, have been developed in multi-kW size for powering electric vehicles. A problem with using an alkaline electrolyte is that it absorbs C 0 2 from the air, poisoning the electrolyte and reducing its conductivity. So the incoming air must be pre-scrubbed to remove CO2. This problem is avoided in the polymer electrolyte membrane fuel cell (PEM) in which the membrane is acidic. You met such membranes in the discussion of the electrolytic chlor-alkali industry in Section 24.1.1 of the main text. A typical membrane consists of a perfluorosulfonic acid polymer containing S 0 2 0 H groups. The -S020H group is strongly acidic, and so allows the passage of hydrogen ions. That this is essential can be seen from the reactions at the two 257
Table 7.1 Principal types of fuel cell
electrodes :
negative electrode 2H2(g) = 4H+(mem) + 4e-
(7.1)
positive electrode O2(g) + 4H+(mem) + 4e- = 2H20(1) overall reaction 2H2(g) + 02(g) = 2 H 2 W
(7.3)
The two electrodes consist of carbon containing pores or channels, which allow passage of the reacting gases. The catalyst that promotes the two electrode reactions is platinum, and it is applied to the carbon in regions where carbon, gas-filled pores and proton-exchange membrane electrolyte all meet (Figure 7.2).
Figure 7.2 Diagram of a PEM fuel cell in which hydrogen is the fuel, and air is the source of oxygen. 258
The PEM fuel cell is one of the more promising types for powering electric vehicles, but the rather high cost of the perfluorosulfonic acid polymer membranes is a limiting factor. In No.rthAmerica, trials are proceeding with city electric buses powered by PEM fuel cell stacks. In Europe, Daimler-Benz have demonstrated their small A-class car in an electric version powered by a PEM fuel cell. This is the NECAR 3, where NE stands for ‘no emissions’ (Figure 7.3a). The pure hydrogen fuel is manufactured on-board by reforming methanol. In this case, reforming generates the greenhouse gas C02, so ‘no emissions’ is not strictly correct. However, this is not true of NECAR 4 (Figure 7.3b), in which the fuel is stored hydrogen. Rapid advances are being made in this technology and there is optimism that fuel-cell-powered cars will meet the requirements for a viable electric vehicle. There are two high-temperature fuel cells, the molten carbonate fuel cell (MCFC) and the solid oxide fuel cell (SOFC). Both present difficult problems for materials science. Molten alkali carbonate at 650 O C is a most aggressive medium, and corrosion problems are severe in this fuel cell. The SOFC operates at even higher temperatures, in the range 700-1 000 O C , depending on the composition of the solid oxide electrolyte employed. The conduction mechanism is the movement of oxide ions (02-) through a defective oxide lattice. Here the problems are concerned with fabrication of ceramic shapes and the thermal expansion of components. There are also engineering problems relating to heat and mass transfer with both types of high-temperature fuel cell. Nevertheless, considerable research progress has been made, and prototypes of both the MCFC and the SOFC in the multi-kW range have been built and tested. Fuel cell development requires the collaboration of physical chemists, materials scientists and chemical engineers. Good progress is now being made, and there is confidence that fuel cells of all five types can be manufactured that will operate well. When and where they will be introduced for terrestrial applications is now largely a matter of economics.
Figure 7.3 Two Daimler-Benz demonstrator cars powered by fuel cells. NECAR 3 (a) uses methanol as a fuel. It is converted to hydrogen in a catalytic reformer in the rear of the vehicle. In NECAR 4 (b), the fuel is liquid hydrogen, stored in an insulated tank in the same location.
259
Grateful acknowledgement is made to the following sources for permission to reproduce material in this book:
Figure 3.1: Colin Cuthert/Science Photo Library; Figure 5.4: courtesy of the Italian Navy; Figure 5.5: Charles D. Winters/Science Photo Library; Figure 5.6a: courtesy of Sony UK Limited; Figures 5.6b and 5.9: courtesy of Braun Oral Care; Figure 6.1: 0NASDA; Figure 7.3a and b: courtesy of Ballard Power Systems. Every effort has been made to trace all the copyright owners, but if any has been inadvertently overlooked, the publishers will be pleased to make the necessary arrangements at the first opportunity.
260
Note Principal references are given in bold type; picture and table references are shown in italics.
Abrahamson, John, 108 absolute entropies, 79-82 acid rain, 193 acids, enthalpies of neutralization of, 57-60 reactions of, with metals, 18, 20-2 activity series, 37-40 see also metals, ease of oxidation of adiabatically enclosed systems, 55 alkali metal complexes, 176-82,183 alkali metal halides, 160-1 alkali metal hydrides, 167-8 alkali metal nitrides, 168 alkali metal normal oxides, 161-6 alkali metal peroxides, 162-6 alkali metal superoxides, 162-6 alkali metals, 148-52 extraction of, from ores, 27 formation of anions by, 180-2, 183 formation of cations by, 150-1 reactions of, with acids, 20-2 solutions of, in aminoethane, 182 solutions of, in liquid ammonia, 152 see also rubidium alkaline batteries, 244,247, 248, 25 1-2 alkaline earth metal carbonates, 197-8 alkaline earth metal complexes, 200-1 alkaline earth metal halides, dihalides, 198-200 monohalides unknown, 188-91 alkaline earth metal hydroxides, 184, 197 alkaline earth metal oxides, 184, 197 alkaline earth elements, 184, 201 alkaline earth metals, 185-8 extraction of, from ores, 27 reactions of, with water, 187 with acids, 22 alkaline manganese dioxide cells, 247, 248
aluminium, 32-6 extraction of, from ores, 32-5, 36 ionization energies of, 188 physical properties of, 150 production figures for, 35-6 reactions of, with bromine, 42 with oxygen, 99-100 aluminium-silver oxide cells, 250 amalgams, 100 aminoethane, alkali metal solutions in, 180-2 ammonia, alkali metal solutions in liquid, 15 1--2 as ligand, 172-3, 174, 201 synthesis of, 98, 99 ampere hour (Ah) (unit), 244, 245 amphoteric oxideshydroxides, 32, 197 analogous reactions, 112 anode, 239
ball lightning, 106-8 barium, 184 physical properties of, 187 reactions of, with acids, 22 with water, 187 barium carbonate, decomposition of, 198 barium peroxide, 197 bases, enthalpies of neutralization of, 57-60 batteries, 140, 203, 240-59 degradation modes in, 255 discharge and charge of, 243-5 internal resistance of, 244 polarization losses in, 244 sizes and applications of, 241-2 specific energies of, 250-2 specification of, 246-54 see also fuel cells bauxite, 32 beryl, 184 26 I
beryllium, 184 physical properties of, 187 beryllium complexes, 200-1 beryllium halides, 199-200, 20 1 beryllium hydroxide, 197 beryllium oxide, 197, 201 bidentate ligands, 174, 175 bond dissociation energies (D), 123, 136-7 Born, Max, 132 Born exponent, 132, 134 Born-Haber cycle, 122-6, 129, 136-7, 188-9 Born-Land6 equation, 133-5 bromine, reaction of, with aluminium, 42 bronze, 17
caesium, 114, 149-5 1 physical properties of, 149-50 reactions of, with acids, 20-1 with oxygen and halogens, 25 caesium chloride, structure of, 160-1 caesium superoxide, 162 calcium, 184 physical properties of, 187 reactions of, with acids, 22 with water, 187 calcium carbonate, 184, 18.5 decomposition of, 198 neutralization reaction, 184, 192-3 see also lime; limestone calcium chloride, ionic model applied to, 129 lattice energy of, 134, 136 calcium fluoride crystals, 133 calcium hydrogen carbonate, 195 calcium hydroxide, 197 see also slaked lime calcium monochloride, Born-Haber cycle for, 189-90 calcium oxide, 197 reaction of, with silica, 193-4 see also quicklime calibration experiments, 59 calorimeters, 56, 60 calorimetry, 56-60 carbon, as reducing agent, 114, 118-19 see also graphite 262
cassiterite, 3 1 cathode, 239 cation-exchange membrane, 157 CD-ROM information, 266 chemical changes; see reactions chemical equilibria, 39, 41 chlor-alkali industry, 154-7 chlorine, absolute entropy of, 80-1 entropy changes with temperature, 73-9 manufacture of, 30, 155-7 reactions of, with metals, 25, 109, 222-3 with sodium, 41, 61 cinnabar, 28 Clausius, Rudolf, 45 close-packed structures, 13, 15 complexes, 170-5 of alkali metals, 176-82, 183 of alkaline earth metals, 200-1 coordinating atoms, 173, 174 copper, 27 extraction of, from ores, 119 reactions of, with aqueous ions, 24 with zinc sulfate, 23 unreactivity of, with acids, 21-2, 63, 64, 913-9 copper sulfate, 21 reaction of, with zinc, 23 copper sulfate pentahydrate, 170-2 reaction of, with ammonia, 172-3 Coulombic energy, 130-1 crown ethers, 176-9 cryolite (solvent for aluminium extraction), 3 3 4 cryptands, 179,180, 182,183
decomposition temperature; see thermodynamic decomposition temperature Deville, H. E. St-C., 33 Dinniss, James, 108 discharge mode (electrochemical cell), 140, :!38-9, 243-5 disproportionation, 191 dolomite, 184 dolomitic lime, 192 driven cells, 237, 238-9 ductile materials, 13, 15 Dye, James, 182
edta; see ethylenediaminetetraacetate anion Einstein, Albert, 45 electrical conductivity of metals, 15-16 electrical work (wel),53, 203 electrochemical cells, 139-42, 237-9 discharge and charge of, 243-5 see also batteries electrochemical series, 94, 95 electrodes, terminology for, 239 electrolysis, 27, 237, 238-9 extraction of alkaline earth metals by, 185-6 extraction of aluminium by, 32-5 production of sodium hydroxide and chlorine by, 155-7 electrolytic cells; see driven cells electromotive force (e.m.f., E ) , 140, 243 electron affinity (E), 123 Ellingham diagrams, 117- 18 emeralds, 184 en; see ethylenediamine endothermic changes, 42-4 enthalpy changes of, 56 energy conservation; see first law of thermodynamics enthalpies of reaction (AH), 42-4 see also standard enthalpies of reaction enthalpy, 42 enthalpy changes (AH), 42,203 measuring, 56-60 for pure substances, 52-5 see also molar enthalpy changes; standard molar enthalpy changes enthalpy of vaporization, measuring, 56 entropy (S), 4546-7 absolute, 79-82 see also entropy changes entropy changes (AS), 45-5 1 determining, 7 1 for pure substances, 87, 88 for phase transitions, 71-3, 80-1 for reactions, 82-3 variation of, with temperature, 73-9, 115-16 see also standard molar entropy changes equilibria; see chemical equilibria equilibrium constants, 39-40, 204 and Gibbs function changes, 85-6 logarithms of, 86 ethylenediamine (en), as ligand, 173, 174,175 ethylenediaminetetraacetate anion (edta), as ligand, 174, 175, 200
exothermic changes, 42-4,49-50 enthalpy changes of, 57-60
first ionization energies (Zl), 123, 125-6, 189 trends across a Period, 143-5 trends down a Group, 146 first law of thermodynamics, 52-5,72, 203 fluorinated graphite, 249 fuel cells, 256-9
gas constant ( R ) , 85 Gibbs, Josiah Willard, 84 Gibbs function, 84,203 and the equilibrium constant, 85-6, 204 see also standard Gibbs function changes glass, 194 molecular structure of, 195 gold, 13, 14, 17, 27 reactions of, with oxygen and halogens, 25 resistance to oxidation of, 114 unreactivity of, with acids and oxygen, 21-2 graphite, 16 fluorinated, 249 Group I metals; see alkali metals Group I1 metals; see alkaline earth metals Grove, Sir William, 256
Haber process; see ammonia, synthesis o f halide ions, as ligands, 173, 174 Hall, Charles, 33-4 heat, 52 direction of flow of, 47-8 heat capacity, 79 HCroult, Paul, 33 Hew’s law, 63-4, 165 Hund’s rule, 145 hydrates, 170-2, 176, 201
integration, 79 iodine, vaporization of, 66, 137 ionic interactions, 130-2 ionic model, 129 ionic radius, and lattice stability, 165-6
263
ionization energies, first, 123, 125-6, 189 trends across a Period, 143-5 trends down a Group, 146 second, of alkali metals, 149 third, of alkaline earth metals, 188 iron, extraction of, from ores, 119, 194 reactions of, with acids, 18, 20, 22 isolated systems, 45-6
Kapustinskii equation, 135, 136, 165, 190 kinetic stability, 98-100
lanthanum, 97 lattice energies (L), 123-5, 129-37 and alkaline earth metal chloride stability, 189-90 and alkali metal hydride stability, 168 and alkali metal oxide stability, 164-6 Lavoisier. Antoine, 29-30 lead, extraction of, from oxide, 27, 114 lead-acid batteries, 24 1, 250 Leclanche' cells; see zinc-carbon cells Lehn, Jean-Marie, 179 ligands, 1 7 3 4 , 1 7 5 lime, 192-5,196 limelight, 66 limestone, 184, 185, 195, 196 lithium, 149 physical properties of, 149 reactions of, with acids, 20-1, 102-3 lithium fluoride, lattice energy of, 134, 136 lithium-ion batteries, 252-4 lithium oxide, 161, 163 lithium peroxide, 163, 164-5 lithium primary batteries, 248-9, 250, 252 lithium-manganese oxide batteries, 254
Madelung constant, 133, 134 magnesium, 184 extraction of, from ores, 186 ionization energies of, 187, 188 physical properties of, 150, 187
264
reactions of, with acids, 18, 20, 21, 22, 92, 93-4 with aqueous ions, 24 with oxygen, 18-19, 111, 112, 115 with water, 187 magnesium carbonate, decomposition of, 198 magnesium hydroxide, 197 magnesium oxide, 197 malleable materials, 13, 15 MCFC (molten carbonate fuel cell), 258, 259 membrane cell, 156-7 mercury, 28-3 1 extraction of, from sulfide ores, 28 production figures for, 3 1 mercury oxide, decomposition of, 117-1 8 metal complexes; see complexes metallic bonding, 15, 150 metals, ease of oxidation of, 37-40,92-6, 102-3, 104-6 extraction of, from ores, 27-35, 114, 119-20, 126 physical properties of, 13-16 reactions of, 17-26 oxidation and reduction, 18-20 with aqueous H+, 20-2,92-6 with aqueous metal ions, 23-4, 38 with oxygen and halogens, 25-6 Minamata Bay disaster, 30 molar enthalpy changes (AHm),60-2 molecular recognition, I79 molten carbonate fuel cell (MCFC), 258, 259
name-plate capacity (batteries), 245 neutralization reaction, 57-60 nickel-cadmium batteries, 25 1, 252 nickel-metal hydride batteries, 25 1-2, 254, 255 noble metals, 27, 114 nominal capacity (batteries), 245
oxidation, 18-20, 20-2 see also metals, ease of oxidation of oxidizing agents, 25-6, 141-2
p block elements, 144-5, 146 parallel connection, 240 Pedersen, Charles, 176-7 PEM fuel cell, 257-9 Periodic Table, 14 permittivity of vacuum, 130 phase transitions, 56-7 enthalpy changes for, 57 entropy changes for, 7 1-3, 80-1 phosphoric acid fuel cell, 257-8 plutonium, 97 polarization losses (electrochemical cells), 244 polydentate ligands, 174, 175 see also crown ethers; cryptands polymer electrolyte membrane (PEM) fuel cell, 257-9 polytetrafluoroethene (PTFE, teflon), 157 potassium, 149-5 1 physical properties of, 149-50 reactions of, with acids, 20, 102-3 potassium chloride, Born-Haber cycle for, 1 89-90 potassium pennanganate, dissolution of, in organic solvents, 177 potassium superoxide, 162-3 power, 54 primary batteries, 240, 24 1, 247-50 PTFE; see polytetrafluoroethene
quicklime, 192, 193
radium, 185 rated capacity (batteries), 245 reactions, directions of, 48-50 mechanisms of, 99 rates of, 39 see also enthalpies of reaction; entropy changes, for reactions reactivity, 17, 102-3 rechargeable batteries, 24 1 redox potentials, 94 see also standard redox potentials redox reactions, 20, 141-2 reducing agents, 25-6, 141-2 alkali metals as, 152 alkaline earth metals as, 185, 187 reduction, 19-20, 23-4 reference states of elements, 65
repulsive energy, 131 reverberatory furnace, 3 1, 32 reversible cells, 243 rubidium, 17, 149-51 physical properties of, 149reactions of, with acids, 20 rubidium superoxide, 162
salt bridge, 139 sand; see silica second law of thermodynamics, 45-50 secondary batteries, 240,241, 246, 250-4 self-driving cells, 237-8 separators (battery cells), 240 series connection, 240 silica, reaction of, with calcium oxide, 193-4 silicon, reaction of, with oxygen, 108 silver, as noble metal, 27, 114 as reducing agent, 91 reaction of, with oxygen, Gibbs function variations in, 116-17 unreactivity of, with acids, 21-2, 92, 93-4, 102-3 silver chloride, Born-Haber cycle for, 122-5 silver fluoride, Born-Haber cycle for, 125 silver oxide, decomposition of, 117 slaked lime, 192, 193 sodium, as reducing agent, 91 ease of oxidation of, 151 negative ion of, 181-3 physical properties of, 149-50 reactions of, with acids, 20 with chlorine, 41, 61 sodium carbonate, uses of, 154, 158 sodium chloride, 148 Born-Haber cycle for, 122-5 electrolysis of, 155-7 enthalpy change for decomposition of, 63 ionic model of, 129 lattice energy of, 124, 134, 135, 136 structure of, 160 sodium fluoride, Born-Haber cycle for, 125 sodium hydroxide, manufacture of, 30, 155-8 uses of, 154,157 sodium iodide, Born-Haber cycle for, 137 sodium peroxide, 161-2, 163-5 265
sodium-metal-chloride cell, 244,245 SOFC (solid oxide fuel cell), 258, 259 solders, 32 solid oxide fuel cell (SOFC), 259 spontaneous reactions, 41,43 square-planar complexes, 172, 173 stalactites and stalagmites, 195-6 standard e.m.f. (E?), 140 of batteries, 250, 251
standard enthalpies of atomization (AHZ), 123, 125-6 standard enthalpies of formation (A H of aqueous ions, 68-70, 88, 89 of gaseous iodine, 137 of sodium and silver halides, 124-5
F), 65-7, 87, 88-9,96
standard Gibbs function of formation (A G:),
87, 88, 89, 96
standard molar enthalpy changes ( AH,"), 64-70 non-variance of, with temperature, 115- 16 for reactions, 102-3, 1 10- 12 calculating, 67-8 for carbonate decomposition, 197-8 standard molar entropy changes ( AS,"), effect of temperature on, 115-1 6 for reactions, 82-3, 110-12 sign and magnitude of, 112 standard molar Gibbs function changes ( AGf ), 84-6, 8 9 , 9 1 4 , 95,98-9, 102-3, 110-12 and equilibrium constant, 85-6 for cell reactions, 140-2 variation of, with temperature, 115-20 standard redox potentials (p), 141-2 state functions, 55 enthalpy as, 55 entropy as, 71 state of a system, 55 storage capacity (cells, batteries), 245 stored energy delivered (cells, batteries), 245 strontium, 184 physical properties of, I87 reactions of, with acids, 22 with water, 187 strontium carbonate, decomposition of, 198 systems, 45
266
teflon; see polytetrafluoroethene thermodynamic decomposition temperature, 116-1 7, 1 6 3 4 thermodynamicstability, 98-1 00, 190-1 thermodynamics, 13,203-4 see also first, second and third laws of therinodynamics (and speclfic thermodynamics functions) third law of thermodynamics, 79-80 Thomsen, Julius, 43,#4 Thomsen's hypothesis, 42-4, 110 tin, 31-2 extraction of, from oxide, 3 1-2 production figures for, 32 reactions of, with acids, 18, 20
units, of electrical conductivity, 15 of entropy, 47 of power, 54 uranium, 17
voltaic cells; see self-driving cells
water, enthalpy change of vaporization of, 56,60, 72-3 entropy change of vaporization of, 72-3 as ligand, 171-2, 174, 201 watt (unit), 54 watt hour (W h), 241,245 work (w),53,203
zinc, reactions of, with acids, 18, 20, 22 with oxygen, 110-1 1 , 112 zinc-carbon cells, 247, 248 zinc-chlorine cell, 237-9, 240
Note Principal references are given in bold type; picture and table references are shown in italics.
Abrahamson, John, 108 absolute entropies, 79-82 acid rain, 193 acids, enthalpies of neutralization of, 57-60 reactions of, with metals, 18, 20-2 activity series, 37-40 see also metals, ease of oxidation of adiabatically enclosed systems, 55 alkali metal complexes, 176-82,183 alkali metal halides, 160-1 alkali metal hydrides, 167-8 alkali metal nitrides, 168 alkali metal normal oxides, 161-6 alkali metal peroxides, 162-6 alkali metal superoxides, 162-6 alkali metals, 148-52 extraction of, from ores, 27 formation of anions by, 180-2, 183 formation of cations by, 150-1 reactions of, with acids, 20-2 solutions of, in aminoethane, 182 solutions of, in liquid ammonia, 152 see also rubidium alkaline batteries, 244,247, 248, 25 1-2 alkaline earth metal carbonates, 197-8 alkaline earth metal complexes, 200-1 alkaline earth metal halides, dihalides, 198-200 monohalides unknown, 188-91 alkaline earth metal hydroxides, 184, 197 alkaline earth metal oxides, 184, 197 alkaline earth elements, 184, 201 alkaline earth metals, 185-8 extraction of, from ores, 27 reactions of, with water, 187 with acids, 22 alkaline manganese dioxide cells, 247, 248
aluminium, 32-6 extraction of, from ores, 32-5, 36 ionization energies of, 188 physical properties of, 150 production figures for, 35-6 reactions of, with bromine, 42 with oxygen, 99-100 aluminium-silver oxide cells, 250 amalgams, 100 aminoethane, alkali metal solutions in, 180-2 ammonia, alkali metal solutions in liquid, 15 1--2 as ligand, 172-3, 174, 201 synthesis of, 98, 99 ampere hour (Ah) (unit), 244, 245 amphoteric oxideshydroxides, 32, 197 analogous reactions, 112 anode, 239
ball lightning, 106-8 barium, 184 physical properties of, 187 reactions of, with acids, 22 with water, 187 barium carbonate, decomposition of, 198 barium peroxide, 197 bases, enthalpies of neutralization of, 57-60 batteries, 140, 203, 240-59 degradation modes in, 255 discharge and charge of, 243-5 internal resistance of, 244 polarization losses in, 244 sizes and applications of, 241-2 specific energies of, 250-2 specification of, 246-54 see also fuel cells bauxite, 32 beryl, 184 26 I
beryllium, 184 physical properties of, 187 beryllium complexes, 200-1 beryllium halides, 199-200, 20 1 beryllium hydroxide, 197 beryllium oxide, 197, 201 bidentate ligands, 174, 175 bond dissociation energies (D), 123, 136-7 Born, Max, 132 Born exponent, 132, 134 Born-Haber cycle, 122-6, 129, 136-7, 188-9 Born-Land6 equation, 133-5 bromine, reaction of, with aluminium, 42 bronze, 17
caesium, 114, 149-5 1 physical properties of, 149-50 reactions of, with acids, 20-1 with oxygen and halogens, 25 caesium chloride, structure of, 160-1 caesium superoxide, 162 calcium, 184 physical properties of, 187 reactions of, with acids, 22 with water, 187 calcium carbonate, 184, 18.5 decomposition of, 198 neutralization reaction, 184, 192-3 see also lime; limestone calcium chloride, ionic model applied to, 129 lattice energy of, 134, 136 calcium fluoride crystals, 133 calcium hydrogen carbonate, 195 calcium hydroxide, 197 see also slaked lime calcium monochloride, Born-Haber cycle for, 189-90 calcium oxide, 197 reaction of, with silica, 193-4 see also quicklime calibration experiments, 59 calorimeters, 56, 60 calorimetry, 56-60 carbon, as reducing agent, 114, 118-19 see also graphite 262
cassiterite, 3 1 cathode, 239 cation-exchange membrane, 157 CD-ROM information, 266 chemical changes; see reactions chemical equilibria, 39, 41 chlor-alkali industry, 154-7 chlorine, absolute entropy of, 80-1 entropy changes with temperature, 73-9 manufacture of, 30, 155-7 reactions of, with metals, 25, 109, 222-3 with sodium, 41, 61 cinnabar, 28 Clausius, Rudolf, 45 close-packed structures, 13, 15 complexes, 170-5 of alkali metals, 176-82, 183 of alkaline earth metals, 200-1 coordinating atoms, 173, 174 copper, 27 extraction of, from ores, 119 reactions of, with aqueous ions, 24 with zinc sulfate, 23 unreactivity of, with acids, 21-2, 63, 64, 913-9 copper sulfate, 21 reaction of, with zinc, 23 copper sulfate pentahydrate, 170-2 reaction of, with ammonia, 172-3 Coulombic energy, 130-1 crown ethers, 176-9 cryolite (solvent for aluminium extraction), 3 3 4 cryptands, 179,180, 182,183
decomposition temperature; see thermodynamic decomposition temperature Deville, H. E. St-C., 33 Dinniss, James, 108 discharge mode (electrochemical cell), 140, :!38-9, 243-5 disproportionation, 191 dolomite, 184 dolomitic lime, 192 driven cells, 237, 238-9 ductile materials, 13, 15 Dye, James, 182
edta; see ethylenediaminetetraacetate anion Einstein, Albert, 45 electrical conductivity of metals, 15-16 electrical work (wel),53, 203 electrochemical cells, 139-42, 237-9 discharge and charge of, 243-5 see also batteries electrochemical series, 94, 95 electrodes, terminology for, 239 electrolysis, 27, 237, 238-9 extraction of alkaline earth metals by, 185-6 extraction of aluminium by, 32-5 production of sodium hydroxide and chlorine by, 155-7 electrolytic cells; see driven cells electromotive force (e.m.f., E ) , 140, 243 electron affinity (E), 123 Ellingham diagrams, 117- 18 emeralds, 184 en; see ethylenediamine endothermic changes, 42-4 enthalpy changes of, 56 energy conservation; see first law of thermodynamics enthalpies of reaction (AH), 42-4 see also standard enthalpies of reaction enthalpy, 42 enthalpy changes (AH), 42,203 measuring, 56-60 for pure substances, 52-5 see also molar enthalpy changes; standard molar enthalpy changes enthalpy of vaporization, measuring, 56 entropy (S), 4546-7 absolute, 79-82 see also entropy changes entropy changes (AS), 45-5 1 determining, 7 1 for pure substances, 87, 88 for phase transitions, 71-3, 80-1 for reactions, 82-3 variation of, with temperature, 73-9, 115-16 see also standard molar entropy changes equilibria; see chemical equilibria equilibrium constants, 39-40, 204 and Gibbs function changes, 85-6 logarithms of, 86 ethylenediamine (en), as ligand, 173, 174,175 ethylenediaminetetraacetate anion (edta), as ligand, 174, 175, 200
exothermic changes, 42-4,49-50 enthalpy changes of, 57-60
first ionization energies (Zl), 123, 125-6, 189 trends across a Period, 143-5 trends down a Group, 146 first law of thermodynamics, 52-5,72, 203 fluorinated graphite, 249 fuel cells, 256-9
gas constant ( R ) , 85 Gibbs, Josiah Willard, 84 Gibbs function, 84,203 and the equilibrium constant, 85-6, 204 see also standard Gibbs function changes glass, 194 molecular structure of, 195 gold, 13, 14, 17, 27 reactions of, with oxygen and halogens, 25 resistance to oxidation of, 114 unreactivity of, with acids and oxygen, 21-2 graphite, 16 fluorinated, 249 Group I metals; see alkali metals Group I1 metals; see alkaline earth metals Grove, Sir William, 256
Haber process; see ammonia, synthesis o f halide ions, as ligands, 173, 174 Hall, Charles, 33-4 heat, 52 direction of flow of, 47-8 heat capacity, 79 HCroult, Paul, 33 Hew’s law, 63-4, 165 Hund’s rule, 145 hydrates, 170-2, 176, 201
integration, 79 iodine, vaporization of, 66, 137 ionic interactions, 130-2 ionic model, 129 ionic radius, and lattice stability, 165-6
263
ionization energies, first, 123, 125-6, 189 trends across a Period, 143-5 trends down a Group, 146 second, of alkali metals, 149 third, of alkaline earth metals, 188 iron, extraction of, from ores, 119, 194 reactions of, with acids, 18, 20, 22 isolated systems, 45-6
Kapustinskii equation, 135, 136, 165, 190 kinetic stability, 98-100
lanthanum, 97 lattice energies (L), 123-5, 129-37 and alkaline earth metal chloride stability, 189-90 and alkali metal hydride stability, 168 and alkali metal oxide stability, 164-6 Lavoisier. Antoine, 29-30 lead, extraction of, from oxide, 27, 114 lead-acid batteries, 24 1, 250 Leclanche' cells; see zinc-carbon cells Lehn, Jean-Marie, 179 ligands, 1 7 3 4 , 1 7 5 lime, 192-5,196 limelight, 66 limestone, 184, 185, 195, 196 lithium, 149 physical properties of, 149 reactions of, with acids, 20-1, 102-3 lithium fluoride, lattice energy of, 134, 136 lithium-ion batteries, 252-4 lithium oxide, 161, 163 lithium peroxide, 163, 164-5 lithium primary batteries, 248-9, 250, 252 lithium-manganese oxide batteries, 254
Madelung constant, 133, 134 magnesium, 184 extraction of, from ores, 186 ionization energies of, 187, 188 physical properties of, 150, 187
264
reactions of, with acids, 18, 20, 21, 22, 92, 93-4 with aqueous ions, 24 with oxygen, 18-19, 111, 112, 115 with water, 187 magnesium carbonate, decomposition of, 198 magnesium hydroxide, 197 magnesium oxide, 197 malleable materials, 13, 15 MCFC (molten carbonate fuel cell), 258, 259 membrane cell, 156-7 mercury, 28-3 1 extraction of, from sulfide ores, 28 production figures for, 3 1 mercury oxide, decomposition of, 117-1 8 metal complexes; see complexes metallic bonding, 15, 150 metals, ease of oxidation of, 37-40,92-6, 102-3, 104-6 extraction of, from ores, 27-35, 114, 119-20, 126 physical properties of, 13-16 reactions of, 17-26 oxidation and reduction, 18-20 with aqueous H+, 20-2,92-6 with aqueous metal ions, 23-4, 38 with oxygen and halogens, 25-6 Minamata Bay disaster, 30 molar enthalpy changes (AHm),60-2 molecular recognition, I79 molten carbonate fuel cell (MCFC), 258, 259
name-plate capacity (batteries), 245 neutralization reaction, 57-60 nickel-cadmium batteries, 25 1, 252 nickel-metal hydride batteries, 25 1-2, 254, 255 noble metals, 27, 114 nominal capacity (batteries), 245
oxidation, 18-20, 20-2 see also metals, ease of oxidation of oxidizing agents, 25-6, 141-2
p block elements, 144-5, 146 parallel connection, 240 Pedersen, Charles, 176-7 PEM fuel cell, 257-9 Periodic Table, 14 permittivity of vacuum, 130 phase transitions, 56-7 enthalpy changes for, 57 entropy changes for, 7 1-3, 80-1 phosphoric acid fuel cell, 257-8 plutonium, 97 polarization losses (electrochemical cells), 244 polydentate ligands, 174, 175 see also crown ethers; cryptands polymer electrolyte membrane (PEM) fuel cell, 257-9 polytetrafluoroethene (PTFE, teflon), 157 potassium, 149-5 1 physical properties of, 149-50 reactions of, with acids, 20, 102-3 potassium chloride, Born-Haber cycle for, 1 89-90 potassium pennanganate, dissolution of, in organic solvents, 177 potassium superoxide, 162-3 power, 54 primary batteries, 240, 24 1, 247-50 PTFE; see polytetrafluoroethene
quicklime, 192, 193
radium, 185 rated capacity (batteries), 245 reactions, directions of, 48-50 mechanisms of, 99 rates of, 39 see also enthalpies of reaction; entropy changes, for reactions reactivity, 17, 102-3 rechargeable batteries, 24 1 redox potentials, 94 see also standard redox potentials redox reactions, 20, 141-2 reducing agents, 25-6, 141-2 alkali metals as, 152 alkaline earth metals as, 185, 187 reduction, 19-20, 23-4 reference states of elements, 65
repulsive energy, 131 reverberatory furnace, 3 1, 32 reversible cells, 243 rubidium, 17, 149-51 physical properties of, 149reactions of, with acids, 20 rubidium superoxide, 162
salt bridge, 139 sand; see silica second law of thermodynamics, 45-50 secondary batteries, 240,241, 246, 250-4 self-driving cells, 237-8 separators (battery cells), 240 series connection, 240 silica, reaction of, with calcium oxide, 193-4 silicon, reaction of, with oxygen, 108 silver, as noble metal, 27, 114 as reducing agent, 91 reaction of, with oxygen, Gibbs function variations in, 116-17 unreactivity of, with acids, 21-2, 92, 93-4, 102-3 silver chloride, Born-Haber cycle for, 122-5 silver fluoride, Born-Haber cycle for, 125 silver oxide, decomposition of, 117 slaked lime, 192, 193 sodium, as reducing agent, 91 ease of oxidation of, 151 negative ion of, 181-3 physical properties of, 149-50 reactions of, with acids, 20 with chlorine, 41, 61 sodium carbonate, uses of, 154, 158 sodium chloride, 148 Born-Haber cycle for, 122-5 electrolysis of, 155-7 enthalpy change for decomposition of, 63 ionic model of, 129 lattice energy of, 124, 134, 135, 136 structure of, 160 sodium fluoride, Born-Haber cycle for, 125 sodium hydroxide, manufacture of, 30, 155-8 uses of, 154,157 sodium iodide, Born-Haber cycle for, 137 sodium peroxide, 161-2, 163-5 265
sodium-metal-chloride cell, 244,245 SOFC (solid oxide fuel cell), 258, 259 solders, 32 solid oxide fuel cell (SOFC), 259 spontaneous reactions, 41,43 square-planar complexes, 172, 173 stalactites and stalagmites, 195-6 standard e.m.f. (E?), 140 of batteries, 250, 251
standard enthalpies of atomization (AHZ), 123, 125-6 standard enthalpies of formation (A H of aqueous ions, 68-70, 88, 89 of gaseous iodine, 137 of sodium and silver halides, 124-5
F), 65-7, 87, 88-9,96
standard Gibbs function of formation (A G:),
87, 88, 89, 96
standard molar enthalpy changes ( AH,"), 64-70 non-variance of, with temperature, 115- 16 for reactions, 102-3, 1 10- 12 calculating, 67-8 for carbonate decomposition, 197-8 standard molar entropy changes ( AS,"), effect of temperature on, 115-1 6 for reactions, 82-3, 110-12 sign and magnitude of, 112 standard molar Gibbs function changes ( AGf ), 84-6, 8 9 , 9 1 4 , 95,98-9, 102-3, 110-12 and equilibrium constant, 85-6 for cell reactions, 140-2 variation of, with temperature, 115-20 standard redox potentials (p), 141-2 state functions, 55 enthalpy as, 55 entropy as, 71 state of a system, 55 storage capacity (cells, batteries), 245 stored energy delivered (cells, batteries), 245 strontium, 184 physical properties of, I87 reactions of, with acids, 22 with water, 187 strontium carbonate, decomposition of, 198 systems, 45
266
teflon; see polytetrafluoroethene thermodynamic decomposition temperature, 116-1 7, 1 6 3 4 thermodynamicstability, 98-1 00, 190-1 thermodynamics, 13,203-4 see also first, second and third laws of therinodynamics (and speclfic thermodynamics functions) third law of thermodynamics, 79-80 Thomsen, Julius, 43,#4 Thomsen's hypothesis, 42-4, 110 tin, 31-2 extraction of, from oxide, 3 1-2 production figures for, 32 reactions of, with acids, 18, 20
units, of electrical conductivity, 15 of entropy, 47 of power, 54 uranium, 17
voltaic cells; see self-driving cells
water, enthalpy change of vaporization of, 56,60, 72-3 entropy change of vaporization of, 72-3 as ligand, 171-2, 174, 201 watt (unit), 54 watt hour (W h), 241,245 work (w),53,203
zinc, reactions of, with acids, 18, 20, 22 with oxygen, 110-1 1 , 112 zinc-carbon cells, 247, 248 zinc-chlorine cell, 237-9, 240
The CD-ROMs are designed for use on a PC running Windows 95,98, ME or 2000. We recommend the following as the minimum hardware specification: Pentium 400 MHz or compatible processor memory (RAM) 32 MB hard disk free space 100MB video resolution 800 x 600 pixels at High Colour (16 bit) 8 x CD-ROM CD-ROM speed sound card and speakers Windows compatible
The contents of this CD-ROM have been through many quality control checks at the Open University, and we do not anticipate that you will encounter difficulties in installing and running the software. However, a website will be maintained at http://the-molecular-world.open.ac.uk which records solutions to any faults that are reported to us.
Computers with higher specification components will provide a smoother presentation of the multimedia materials.
Software must be installed onto your computer before you can access the applications. Please run 1NSTALL.EXE from the CD-ROM. This program may direct you to install other, third party, software applications. You will find the installation programs for these applications in the INSTALL folder on the CD-ROMs. To access all the software on these CD-ROMs, you must install QuickTime, Isis/Draw, WebLab ViewerLite and Acrobat Reader.
Figure C.l The CD-ROM Guide.
You can access Metals and Chemical Change CD-ROM applications through a CD-ROM Guide (Figure C. l), which is created as part of the installation process. You may open this from the Start menu, by selecting Programs followed by The Molecular World. The CD-ROM Guide has the same title as this book. The Data Book is accessed directly from the Start I Programs I The Molecular World menu (Figure C.2), and is supplied as Adobe Acrobat document. If you are unfamiliar with Acrobat, please run the introduction supplied in the CD-ROM Guide. 272
Figure C.2 Accessing the Data Book and CD-ROM Guide.